Traditional Training Approaches Research Articles

Training of food handlers using a virtual reality environment on handwashing is more effective than traditional training approaches in a randomized trial

ABSTRACT Handwashing is an essential prevention step in foodservice, This randomized comparative study assessed the effectiveness of three training methods -virtual reality (VR), Glo-Germ (GT), and traditional face-to-face training (CT)- in improving handwashing practices among foodservice workers (n = 129). Training focused on the World Health Organization’s 11-step handwashing process. Handwashing in the workplace was observed, in a pre- and post-test design, with microbiological swabs also taken. There was an increase in time spent on handwashing that was noticeable across all three groups (p < 0.05) with the highest in the VR group (50.16 ± 26.04 sec.). The mean number of handwashing steps increased across all groups, and the highest increase (8.91 ± 1.49 steps) was observed in the VR group. A statistically significant difference was observed among all groups regarding correctly performed steps, and the VR group showed the highest number of steps (p < 0.01). These findings suggest that VR-based training has value for enhancing hand hygiene practices in foodservice settings.

Universal Knowledge Graph Embedding Framework Based on High-Quality Negative Sampling and Weighting

The traditional model training approach based on negative sampling randomly samples a portion of negative samples for training, which can easily overlook important negative samples and adversely affect the training of knowledge graph embedding models. Some researchers have explored non-sampling model training frameworks that use all unobserved triples as negative samples to improve model training performance. However, both training methods inevitably introduce false negative samples and easy-to-separate negative samples that are far from the model’s decision boundary, and they do not consider the adverse effects of long-tail entities and relations during training, thus limiting the improvement of model training performance. To address this issue, we propose a universal knowledge graph embedding framework based on high-quality negative sampling and weighting, called HNSW-KGE. First, we conduct pre-training based on the NS-KGE non-sampling training framework to quickly obtain an initial set of relatively high-quality embedding vector representations for all entities and relations. Second, we design a candidate negative sample set construction strategy that samples a certain number of negative samples that are neither false negatives nor easy-to-separate negatives for all positive triples, based on the embedding vectors obtained from pre-training. This ensures the provision of high-quality negative samples for model training. Finally, we apply weighting to the loss function based on the frequency of the entities and relations appearing in the triples to mitigate the adverse effects of long-tail entities and relations on model training. Experiments conducted on benchmark datasets FB15K237 and WN18RR using various knowledge graph embedding models demonstrate that our proposed framework HNSW-KGE, based on high-quality negative sampling and weighting, achieves better training performance and exhibits versatility, making it applicable to various types of knowledge embedding models.

Development of Virtual Reality-Based Left Brain System

This paper proposes a virtual reality based Left Brain System named BrainUp World to improve left brain thinking. The left-brain system was employed with mobile virtual reality technology, hand motion tracking and haptic feedback system. The implementation of these systems is to enhance the experience and sense of embodiment. BrainUp World includes virtual reality-based brain training games to improve a person’s attention level, reasoning skill, auditory cognition, arithmetic skill, sequencing skill and memory. The hand tracking system was utilized with an IR camera to capture the hand orientation and return the gesture data to the VR application. A low-cost and lightweight haptic glove was invented which provides the sense of touch using vibrations while interacting with VR contents. An experimental study was conducted to assess the efficacy of BrainUp World compared to traditional PC-based training approaches. Participants were randomly assigned to either the VR-based or PC-based training group and underwent 6 different games to test a person’s attention level, reasoning skill, auditory cognition, arithmetic skill, sequencing skill, and memory. The results revealed a statistically significant improvement in left brain function in the VR-based training group compared to the PC-based group, with a T-score growth of 4.73. Analysis using ANOVA confirmed the significance of this difference (p-Value = 0.04). Notably, the study also identified age-related differences in thinking fluency, highlighting the importance of personalized cognitive training interventions. In conclusion, BrainUp World demonstrates the potential of VR technology in promoting left brain development, as evidenced by empirical findings from the conducted study. By offering immersive and interactive cognitive training experiences, VR-based systems hold promise for enhancing various aspects of cognitive function associated with left hemisphere dominance. Further research in this area is encouraged to explore the full potential of VR-based interventions for cognitive enhancement.

Improving Machine Learned Force Fields for Complex Fluids through Enhanced Sampling: A Liquid Crystal Case Study.

Machine learned force fields offer the potential for faster execution times while retaining the accuracy of traditional DFT calculations, making them promising candidates for molecular simulations in cases where reliable classical force fields are not available. Some of the challenges associated with machine learned force fields include simulation stability over extended periods of time and ensuring that the statistical and dynamical properties of the underlying simulated systems are correctly captured. In this work, we propose a systematic training pipeline for such force fields that leads to improved model quality, compared to that achieved by traditional data generation and training approaches. That pipeline relies on the use of enhanced sampling techniques, and it is demonstrated here in the context of a liquid crystal, which exemplifies many of the challenges that are encountered in fluids and materials with complex free energy landscapes. Our results indicate that, whereas the majority of traditional machine learned force field training approaches lead to molecular dynamics simulations that are only stable over hundred-picosecond trajectories, our approach allows for stable simulations over tens of nanoseconds for organic molecular systems comprising thousands of atoms.

The Autoregulation Rest-Redistribution Training Method Mitigates Sex Differences in Neuromuscular and Perceived Fatigue During Resistance Training.

To examine the sex differences in performance and perceived fatigue during resistance training prescribed using traditional (TRA) and autoregulation rest-redistribution training (ARRT) approaches. Twelve resistance-trained men and 12 women completed 2 sessions including the bench-press exercise matched for load (75% of 1-repetition maximum), volume (24 repetitions), and total rest (240s). Sessions were performed in a counterbalanced randomized design with TRA consisting of 3 sets of 8 repetitions with 120-second interset rest and ARRT employing a personalized combination of clusters, repetitions per cluster, and between-clusters rest regulated with a 20% velocity-loss threshold. The effects of TRA and ARRT on velocity loss, unilateral isometric peak force, and rating of fatigue (ROF) were compared between sexes. The velocity loss was generally lower during ARRT compared with TRA (-0.47% [0.11%]), with velocity loss being mitigated by ARRT to a greater extent among males compared with females (-0.37% [0.15%]). A smaller unilateral isometric peak force decline was observed after ARRT than TRA among males compared with females (-38.4 [8.4]N). Lower ROF after ARRT than TRA was found among males compared to females (-1.97 [0.55]AU). Additionally, males reported greater ROF than females across both conditions (1.92 [0.53]AU), and ARRT resulted in lower ROF than TRA overall (-0.83 [0.39]AU). The ARRT approach resulted in decreased velocity loss, peak force impairment, and ROF compared with TRA in both sexes. However, male subjects exhibited more pronounced acute within-session benefits from the ARRT method.

Enhancing clinical skills in pediatric trainees: a comparative study of ChatGPT-assisted and traditional teaching methods

BackgroundAs artificial intelligence (AI) increasingly integrates into medical education, its specific impact on the development of clinical skills among pediatric trainees needs detailed investigation. Pediatric training presents unique challenges which AI tools like ChatGPT may be well-suited to address.ObjectiveThis study evaluates the effectiveness of ChatGPT-assisted instruction versus traditional teaching methods on pediatric trainees’ clinical skills performance.MethodsA cohort of pediatric trainees (n = 77) was randomly assigned to two groups; one underwent ChatGPT-assisted training, while the other received conventional instruction over a period of two weeks. Performance was assessed using theoretical knowledge exams and Mini-Clinical Evaluation Exercises (Mini-CEX), with particular attention to professional conduct, clinical judgment, patient communication, and overall clinical skills. Trainees’ acceptance and satisfaction with the AI-assisted method were evaluated through a structured survey.ResultsBoth groups performed similarly in theoretical exams, indicating no significant difference (p > 0.05). However, the ChatGPT-assisted group showed a statistically significant improvement in Mini-CEX scores (p < 0.05), particularly in patient communication and clinical judgment. The AI-teaching approach received positive feedback from the majority of trainees, highlighting the perceived benefits in interactive learning and skill acquisition.ConclusionChatGPT-assisted instruction did not affect theoretical knowledge acquisition but did enhance practical clinical skills among pediatric trainees. The positive reception of the AI-based method suggests that it has the potential to complement and augment traditional training approaches in pediatric education. These promising results warrant further exploration into the broader applications of AI in medical education scenarios.

Training Law Enforcement Officers About Autism: Evaluation of Adding Virtual Reality or Simulation to a Traditional Training Approach.

Prevalence rates of autism spectrum disorder (ASD) across the globe and lifespan suggest that law enforcement officers (LEOs) are likely to interact with autistic people in their professional duties. In developed countries, LEOs have professional encounters with autistic individuals, and research strongly supports LEOs need training to increase their knowledge about autism and confidence interacting with autistic people. The purpose of the present study is to compare three training approaches with LEOs: (a) a training presentation alone (P), (b) a training presentation combined with virtual reality simulation (P + VR), and (c) a training presentation combined with in-vivo simulation (P + SIM). The primary outcomes are LEO knowledge of autism and self-reported confidence in responding to autism calls. The present study included 259 LEOS who completed a survey assessing autism knowledge and confidence in level of skill before and after participating in ASD-specific training for LEOs. Findings support that participant knowledge of autism and confidence in responding significantly increased for all LEOs equally across training conditions. Results support that formalized training in ASD significantly impacts LEO's knowledge of ASD and confidence responding to calls. As such, standardized training for LEOs is needed to assure appropriate knowledge and confidence when responding to calls involving autistic persons.

Enhancing technical proficiency through small-sided basketball games: A strategic approach for students athletes

Background Problems: Due to the intricate nature of the game and the requirement for proficiency in core skills like dribbling, passing, shooting, and defensive positioning, basketball players frequently have difficulties developing their technical abilities. Traditional training approaches may not focus enough on these skills, leaving room for growth. Research Objective: This study investigates how well basketball players' technical skills may be improved through game-based instruction, particularly small-sided basketball games. The study aims to address the need for a more focused approach to technical elements by highlighting the value of focused practice in pertinent game scenarios. Methods: The type of research used was quantitative research with a sample of 12 amateur athletes. The instrument used is the game performance evaluation tool (GPET). Quantitative data from pre- and post-assessments were examined using the appropriate statistical techniques, such as paired t-tests or analysis of variance (ANOVA). For the study, basketball players with various skill levels were listed. Small groups of participants received specialised training to develop their technical abilities. Findings and Results: Participants' technical proficiency significantly improved after strategically focused practice using small-sided basketball games. The outcomes show that this game-based method works well for improving basic abilities, including passing, dribbling, shooting, and positioning defensively. The results highlight how effective it is to use small-sided games to help basketball players enhance their skills. Conclusion: The study concludes that small-sided basketball games are an effective strategy for improving basketball players' technical proficiency. Notwithstanding several constraints, such as a comparatively limited sample size and consistency in participant skill levels, the findings indicate that focused practice in pertinent gaming environments produces notable enhancements. To better understand the potential of this strategy for players of different skill levels, future studies should include additional variables like training amounts and differences in small-sided game structures.

Developing the Physical Performance in Youth Soccer: Short-Term Effect of Dynamic-Ecological versus Traditional Training Approach for Sub-Elite U13 Players-An Ecological Exploratory Cluster Randomised Trial.

Currently, research in youth soccer consists of methodological choices that can raise activity volumes and exercise intensity to promote proper training for youth soccer demands. Therefore, the present cluster randomised trial aims to evaluate the effects of the dynamic-ecological approach on the physical performance parameters compared with a traditional one in a group of sub-elite U13 players. Thirty-five male children (age 12.16 ± 0.55 years; weight 45.59 ± 7.15 kg; height 145.5 ± 4.2 cm; BMI 15.8 ± 2.1 kg·m-2) were recruited for this trial from two teams belonging to sub-elite soccer schools and randomly assigned to a dynamic-ecological approach (DEA) or a traditional training (TTG) group. The training program lasted six weeks and consisted of 18 training sessions of 90 min each (3 sessions per week). The sample was evaluated by the standing long jump (SLJ), hop test (HT), 10 m sprint (10 m), 10 × 5 m shuttle run test (SRT), and leg raise test (LR). The DEA group showed significantly higher results in the SLJ (p < 0.001), HT left leg (p < 0.001), 10 m sprint (p < 0.001), and SRT (p < 0.001). In conclusion, the dynamic-ecological approach provides higher performance adaptations. Therefore, this approach can be considered a suitable method to optimise pre-pubertal player training, mainly when no fitness or strength coach is available.

Advancing NLP models with strategic text augmentation: A comprehensive study of augmentation methods and curriculum strategies

This study conducts a thorough evaluation of text augmentation techniques across a variety of datasets and natural language processing (NLP) tasks to address the lack of reliable, generalized evidence for these methods. It examines the effectiveness of these techniques in augmenting training sets to improve performance in tasks such as topic classification, sentiment analysis, and offensive language detection. The research emphasizes not only the augmentation methods, but also the strategic order in which real and augmented instances are introduced during training. A major contribution is the development and evaluation of Modified Cyclical Curriculum Learning (MCCL) for augmented datasets, which represents a novel approach in the field. Results show that specific augmentation methods, especially when integrated with MCCL, significantly outperform traditional training approaches in NLP model performance. These results underscore the need for careful selection of augmentation techniques and sequencing strategies to optimize the balance between speed and quality improvement in various NLP tasks. The study concludes that the use of augmentation methods, especially in conjunction with MCCL, leads to improved results in various classification tasks, providing a foundation for future advances in text augmentation strategies in NLP.

Evolution of HIV Health Care Workforce Needs in the U.S. Mountain West During the COVID-19 Pandemic: A Mixed Method Study.

The COVID-19 pandemic drastically affected health care delivery for vulnerable populations. Many facilities shifted services to telemedicine, and people with HIV or at risk of acquiring HIV experienced interruptions in care. Simultaneously, traditional training approaches to help providers adapt were disrupted. Using a mixed method approach to examine changes over time, we integrated data on trainee needs collected by the Mountain West AIDS Education and Training Center (AETC): a 10-state needs assessment survey in 2020; feedback from a 2020 community of practice; aggregate training data from 2000 to 2022; and a second survey in 2022. HIV care providers' training needs evolved from wanting support on telemedicine and COVID-19 patient care issues, to a later focus on mental health and substance use, social determinants of health, and care coordination. This integrative analysis demonstrates the vital role that AETCs can play in addressing evolving and emergent public health challenges for the HIV workforce.

A novel multi-gradient evolutionary deep learning approach for few-shot wind power prediction using time-series GAN

Due to the lack of historical data, accurate prediction is a great challenge for newly constructed wind farms (NWFs). How to guarantee satisfactory prediction accuracy with limited data has become a bottleneck issue in wind power prediction (WPP). To solve the few-shot learning problem in NWFs, a novel multi-gradient evolutionary deep learning neural network (EATDLNN) prediction model is proposed, which incorporates the time-series GAN (TimeGAN) and multivariate variational mode decomposition (MVMD) method. TimeGAN adds an auto-encoder network and a supervised loss function to GAN, which enables it to capture the dynamic time series correlation characteristics of wind data and generate high-quality samples. MVMD is then employed to simultaneously decompose nonstationary sequences. In the EATDLNN prediction model, an evolution-based multi-gradient training approach is proposed to train the DLNN by integrating three gradient descent methods (i.e., Adam, MBGD-Momentum, RMSprop). Compared with the traditional single-gradient training approach, the multi-gradient training approach provides different gradient descent paths and the gradient can adaptively choose to update towards the path with the smallest error based on the evolutionary framework, which can help the model jump out of the local optimum and enhance the generalization ability. Massive experiments prove the excellent performance of the proposed method, especially in one-step prediction, the root mean square errors reduce by 79.40%∼91.05% compared with other state-of-the-art methods.

An Improved Federated Learning-Assisted Data Aggregation Scheme for Smart Grids

In the context of rapid advancements in artificial intelligence (AI) technology, new technologies, such as federated learning and edge computing, have been widely applied in the power Internet of Things (PIoT). When compared to the traditional centralized training approach, conventional federated learning (FL) significantly enhances privacy protection. Nonetheless, the approach poses privacy concerns, such as inferring other users’ training data through the global model or user-transferred parameters. In light of these challenges, this research paper introduces a novel privacy-preserving data aggregation scheme for the smart grid, bolstered by an improved FL technique. The secure multi-party computation (SMC) and differential privacy (DP) are skillfully combined with FL to combat inference attacks during both the learning process and output inference stages, thus furnishing robust privacy assurances. Through this approach, a trusted third party can securely acquire model parameters from power data holders and securely update the global model in an aggregated way. Moreover, the proposed secure aggregation scheme, as demonstrated through security analysis, achieves secure and reliable data aggregation in the electric PIoT environment. Finally, the experimental analysis shows that the proposed scheme effectively performs federated learning tasks, achieving good model accuracy and shorter execution times.

Effectiveness of nursing virtual training courses based on the Kirkpatrick Model.

Educational evaluation is one of the most significant programs of any organization. In fact, evaluating training effectiveness allows managers and staff of the organization to get a clear picture of the qualities of training activities. The purpose of this study was to evaluate virtual nursing intensive care training courses at Shariati Hospital based on the Kirkpatrick Model in 2020. The current cross-sectional study is an evaluation of the outcome-oriented program to evaluate the results of the e-learning course on nursing intensive care at Shariati Teaching Hospital from March 20, 2020 to September 20, 2020. The total number of nurses working in intensive wards was 168, of whom 150 participated in this research study with convenience sampling. The data collection tool consisted of four-part questionnaires including demographic information and three levels of the Kirkpatrick Model. The mean age of participants was 34.41 ± 6.63 years. The mean score of participant satisfaction was 76.77, indicating that they showed good reaction to virtual training courses. Virtual training course for intensive care nurses was statistically significant at the levels of learning (P < 0.0001) and behavior (P < 0.0001). Virtual training courses are effective educational method based on three levels of Kirkpatrick's Model in intensive care nurses and it can be used along with traditional training approaches.

Procedural learning in virtual reality: The role of immersion, interactivity, and spatial ability.

Although there is much enthusiasm for using virtual reality (VR) in training and education, research findings on the benefits of VR over traditional learning environments are mixed. This disconnect in the literature may be because the cognitive mechanisms that underlie learning in VR have not been addressed systematically. This research explored immersion and interactivity, two key features of VR that set it apart from traditional training approaches. In addition, the impact of spatial ability (SA) on learning outcomes was also assessed. In this experiment, 83 college students learned a maintenance procedure in a virtual environment (VE). Students were assigned randomly to one of three between-subjects training groups. In the Desktop group, students used a desktop-based environment and interacted with it using a mouse and keyboard. In the VR-voice group, students used a VR-based environment and interacted using voice-based commands, while those in the VR-Gesture group interacted using gesture-based commands. To measure learning, students completed a written recall test. There were no significant differences in learning outcome across the three groups, but SA was a significant moderator. Overall, those with high SA outperformed those with low SA in the Desktop condition; however, this effect was not significant in the VR conditions. In fact, those with low SA performed best in the VR-gesture condition compared to the VR-voice or Desktop training. These results suggest that VR environments that incorporate gesture-based interactions may help individuals with low SA better understand the content of the lesson.

H-HOME: A learning framework of federated FANETs to provide edge computing to future delay-constrained IoT systems

In 6G systems, it will be mandatory that the network is able to support edge computing powered by Artificial Intelligence (AI) to provide mobile devices with the opportunity of job offloading for computation, so implementing the new paradigm of Intelligent Internet of Intelligent Things (IIoIT). In areas that are very difficult to be covered by the structured networks, Flying Ad-Hoc Networks (FANET) can be considered one of the most promising technologies to enhance coverage, capacity, reliability, and energy efficiency of wireless cellular networks, also providing edge-computing services. The goal of this paper is to propose a two-layer Hierarchical Horizontal-Offload ManagEment (H-HOME) framework for horizontal offload among the Unmanned Aerial Vehicles (UAV) of the same FANET, in order to minimize processing delay and jitter. The framework exploits Federated Reinforcement Learning in order to take advantage of knowledge sharing without incurring into problems of privacy and network overloading. A Markov Decision Process (MDP) is also defined to optimize decisions of the FANET Orchestrator (FO). Simulation results, obtained from two different analyses, demonstrate that H-HOME outperforms the traditional local training approach, based on simple recursive learning, and it can be effectively used to reduce power consumption and increase FANETs flight autonomy.

Viewpoint-Invariant Loop Closure Detection Using Step-Wise Learning With Controlling Embeddings of Landmarks

The research on loop closure detection has been carried out for the last many years; however, loop closure detection efficiency is still not that good and has been affected by many factors, including illumination conditions, weather conditions, seasons, and viewpoint changes. The research on loop closure detection from a different viewpoint is still an open research problem. The paper proposes an efficient solution to loop closure detection from different viewpoints by using landmarks instead of whole frames and taking the deep and robust features of deep learning instead of handcrafted features. A different kind of training approach is used to train the deep CNN to get highly abstract embeddings of input landmarks. The approach has many advantages over the traditional training approach and can solve many complex problems. This paper has used this approach to force similar or closer embeddings for similar landmarks and is forced to have a large gap in embeddings of different landmarks. The proposed method endeavors viewpoint invariant features, and the astonishing power of deep learning makes the features robust to viewpoint changes, occlusions, and illumination variations. The proposed visual SLAM system is tested on six publicly available datasets, and the results are compared with the most popular Bag of Words methods like DBoW2, DBoW3, and state-of-the-art deep learning methods AlexNet, ResNeXt, FlyNet, AFDPR and Siamese network. The results show that our method is efficient in finding loop closures candidates from different viewpoints. Code is available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/IRMVLab/Step-wise-Learning</uri> .

Immersive virtual reality and passive haptic interfaces to improve procedural learning in a formal training course for first responders

One key aspect for the safety and success of first responders’ operations is the compliance, during the intervention, with all the safety procedures and prescribed behaviors. Although real-world simulation exercises are considered as the best way to verify if operators are ready to handle emergency situations, they are not always a viable approach. Firefighting courses, for example, do not usually include this kind of activities, due to the numerous hazards related to deploying controlled fires for the simulation. However, traditional training approaches based on class lessons and multimedia learning material may not be particularly effective for teaching practical skills and procedural behaviors. In this work, the use of a Virtual Reality Training Simulation (VRTS) combined with passive haptic interfaces and a real-time fire simulation logic is investigated as a complement to a traditional video-based training approach used in the context of forest firefighting. The teaching of safety concepts and correct use of individual firefighting tools was selected as a use case, and a user study involving 45 trainees was carried out in the context of an existing training course. One third of the trainees attended the traditional video-based lessons of the course, whereas the remaining ones also took part to a practice training session, half of them with the devised VRTS, the others in the real world. Experimental results showed that the additional use of the devised VRTS improved the trainees’ procedural learning, as well as their motivation and perceived quality of the overall learning experience.

Methodological characteristics, physiological and physical effects, and future directions for velocity-based training in soccer: A systematic review

Introduction. This systematic review was conducted to (1) characterize the main elements of studies of velocity-based training (VBT) (e.g., training protocols) conducted in soccer, (2) summarize the main physiological and physical effects of VBT on soccer players, and (3) provide future directions for research. Methods: A systematic review of Cochrane Library, EBSCO, PubMed, Scielo, Scopus, SPORTDiscus, and Web of Science databases was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Results: The database search initially identified 127 titles. Of those, five articles were deemed eligible for the systematic review, two studies used a traditional strength training approach, and the other remaining three used sprint training with either resisted sprints or combined resisted and unresisted sprints. All studies addressed strength and power and sprint outcomes, three measured jump performance improvements, and only one study addressed spatiotemporal and kinematics or aerobic measures regarding adaptations to VBT interventions. Only one study addressed acute responses to VBT training regarding spatiotemporal variables and kinematics. Conclusions: Acute responses to VBT training were as follows: when sprint time decreases by at least 50–60%, sprint kinematics are immediately affected, but spatiotemporal variables are only significantly affected when velocity loss (v.loss) reaches at least 60%. For long-term adaptations, it seems that for strength increases using the squat, higher or lower velocity loss due to in-set fatigue accumulation does not make a difference, although it does affect jump performance, favoring the low v.loss groups (15%). The same applies to sprint, as low v.loss accumulation due to fatigue along sets seems to be detrimental to sprint performance adaptations. Moreover, high v.loss during sprints due to external load can improve sprint performance without harming the running technique as was previously thought.

Improving Reading Skills Using a Computerized Phonological Training Program in Early Readers with Reading Difficulties.

In the last years, there has been a big effort to identify risk factors for reading difficulties and to develop new methodologies to help struggling readers. It has been shown that early intervention is more successful than late intervention, and that intensive training programs can benefit children with reading difficulties. The aim of our study is to investigate the effectiveness of an intensive computerized phonological training program designed to improve reading performance in a sample of children with reading difficulties at the early stages of their reading learning process. Thirty-two children with reading difficulties were randomly assigned to one of the two intervention groups: RDIR (children with reading difficulties following a computerized intensive remediation strategy) (n = 20) (7.01 ± 0.69 years), focused on training phonemic awareness, decoding and reading fluency through the computational training; and RDOR (children with reading difficulties following an ordinary remediation strategy) (n = 12) (6.92 ± 0.82 years), which consisted of a reinforcement of reading with a traditional training approach at school. Normal readers (NR) were assigned to the control group (n = 24) (7.32 ± 0.66 years). Our results indicate that both the RDIR and RDOR groups showed an increased reading performance after the intervention. However, children in the RDIR group showed a stronger benefit than the children in the RDOR group, whose improvement was weaker. The control group did not show significant changes in reading performance during the same period. In conclusion, results suggest that intensive early intervention based on phonics training is an effective strategy to remediate reading difficulties, and that it can be used at school as the first approach to tackle such difficulties.