Comparing the efficacy of LORETA neurofeedback versus cognitive rehabilitation on working memory in children with learning disorders: A double-blind clinical trial

Abstract

BACKGROUND:

Nowadays, researchers emphasize the use of inexpensive and low-risk methods as preferred treatments for disorders in children and adolescents. The present study was conducted with the aim of comparing the effects of LORETA neurofeedback (NF) and cognitive rehabilitation on the working memory of children with learning disabilities.

MATERIALS AND METHODS:

The present research design was a quasi-experimental with pretest, posttest with three-month follow-up. The population were all students with learning disorders and sample was 45 children aged between 7 and 10 with learning disabilities from schools in Shiraz. Three groups of 15 children were selected by convenience sampling. The intervention consisted of cognitive rehabilitation and LORETA NF training sessions. To measure outcomes, they used three assessment tools: the IVA (Integrated Visual Auditory Performance Assessment), Wechsler’s Working Memory Test, and the Children’s Clinical Assessment. Data were analyzed by MANCOVA using SPSS 28.

RESULTS:

The study’s results revealed that participants who received LORETA NF treatment showed more significant improvements in their working memory compared to those who underwent cognitive rehabilitation (P < 0.05). This improvement was evident not only immediately after treatment but also persisted through the follow-up period, suggesting that NF had a more lasting positive effect than cognitive rehabilitation (P < 0.05).

CONCLUSION:

The results suggest that LORETA NF training could be an effective intervention for improving working memory in children who have learning disabilities. Therapists and neuropsychologists can use the results of this study to improve the cognitive functions of affected students.

Background

Working memory deficits are widely recognized as a core feature of learning disorders, significantly impacting children’s academic performance and daily functioning.[1,2] As the prevalence of learning disorders continues to rise, reaching approximately 5–15% of school-age children globally, the need for effective interventions targeting working memory has become increasingly crucial.[3,4,5,6] Among the various treatment approaches that have emerged, LORETA NF and cognitive rehabilitation represent two distinct yet promising methodologies, each offering unique mechanisms for enhancing working memory function.[7,8,9,10,11,12]

LORETA NF, a specialized form of biofeedback that enables individuals to modulate their brain activity through real-time monitoring and feedback, has gained considerable attention in recent years.[13,14,15] This approach operates on the premise that children can learn to voluntarily regulate their neural patterns, particularly in regions associated with working memory, such as the prefrontal cortex and posterior parietal areas.[16,17] The process typically involves presenting brain activity patterns through visual or auditory feedback, allowing children to develop strategies for optimizing their neural functioning.[18,19]

In contrast, cognitive rehabilitation encompasses a broader spectrum of targeted interventions designed to strengthen specific cognitive processes through structured training exercises. These programs often employ computerized tasks that systematically challenge working memory capacity, adapting difficulty levels based on performance. The theoretical foundation of cognitive rehabilitation rests on neuroplasticity principles, suggesting that repeated engagement in progressively challenging working memory tasks can lead to sustainable improvements in cognitive function.[20]

Despite the growing implementation of both approaches in clinical settings, comparative analyses of their effectiveness in addressing working memory deficits among children with learning disorders remain limited.[21,22] The existing literature presents mixed findings, with some studies suggesting superior outcomes for LORRETA NF in terms of sustained attention and cognitive flexibility, while others indicate more robust working memory improvements through cognitive rehabilitation protocols.[23,24] This discrepancy highlights the need for a comprehensive examination of both interventions’ relative efficacy, considering factors, such as treatment duration, intensity, and specific learning disorder profiles. Understanding the comparative advantages and limitations of NF and cognitive rehabilitation is essential for developing evidence-based treatment recommendations and optimizing intervention strategies for children with learning disorders. LORETA Neurofeedback is a low-risk and innovative therapy that affects the patient’s micro-neurological dimensions. In fact, the innovation of using this therapy is that this form of neurofeedback has been rarely used in previous studies and its comparison with cognitive rehabilitation has not been made. This review aims to critically evaluate the current evidence regarding the effectiveness of these approaches in enhancing working memory function, examining their underlying mechanisms, practical implications, and potential for complementary implementation in clinical practice.

Materials and Methods

Study design and setting

The present research design was a quasi-experimental with pretest, posttest, and three-month follow-up.

Study participants and sampling

The statistical population of the study was children with learning disabilities in Shiraz who were referred to the counseling center. Statistical indexes, namely the mean and standard deviation, were calculated for three distinct groups. The mean score for group 1 was 2.85, while for groups 2 and 3, they were 2.44 and 2.65. Furthermore, the standard deviations for three groups were determined between to be 0.7 and 0.4. These calculations were based on prior studies and a pilot study conducted by previous authors.[5] To ensure adequate statistical power (α) of 0.05 and a power (β) of 0.2, it was initially determined that a minimum of 15 patients were essential for each group, resulting in a total of 45 patients for both groups (n = 45). However, considering the potential attrition rate, a statistics specialist recommended increasing the sample size. Therefore, it was proposed to recruit 60 patients for each of the three groups, resulting in a total sample size of 60 patients (n = 60). By adopting this larger sample size, it was anticipated that the study would maintain a statistical power of 80% (P < 0.05), ensuring a greater likelihood of detecting meaningful effects or differences between the groups.

The students were between the ages of 7 and 10 and of both genders. Subjects were randomly assigned into three groups of 15 people based on the convenience sampling method, based on the diagnosis and referral of psychiatric specialists, and according to the inclusion and exclusion criteria. Inclusion criteria included receiving a diagnosis of a learning disorder, no history of psychosis or agitation according to the diagnosis of a psychiatrist, age range between 7 and 10 years, not receiving medication or other psychological treatment at the time of diagnosis and consent and declaration of informed and voluntary readiness to participate in the study.

Data collection tool and technique

Wechsler Test: This software was designed by Wechsler in 2003 based on the Wechsler Memory subtest with the ability to measure general verbal comprehension, perceptual reasoning, working memory, processing speed, and general IQ, and can be used to measure the memory of children and adults. The internal reliability coefficient of the Wechsler Working Memory Scale is very high and has a reliability coefficient of more than 90. The test–posttest validity of this task is with a reliability coefficient of 80–89.[25] In Iran, a study conducted by Saed et al.[26] found that the reliability of this test was 0.74 using the Cronbach’s alpha method and 0.75 using the split-half method. Cronbach’s alpha method was used to measure the reliability of this questionnaire. This index was calculated to be equal to 0.91.

Clinical Q Assessment: A method of assessing and determining a treatment protocol for NF. the purpose of using Clinical Q is to make the NF treatment process more effective, using this method reduces the number of NF sessions (Swingle, 2008). The advantage of this method is that in the evaluation and treatment process, harmonies are broadcast and the brain’s response to them is measured, according to the brain’s response to the harmonies, it is used in the treatment sessions to guide the brain to increase or decrease the desired brain wave. There are four harmonies in this process: CERN, Sweep, Mozart, and Omni. The recording process is performed in four stages and is two-channel, two electrodes are installed on the head and two electrodes are installed on each ear. In the first stage, the channel A electrode is on F4 and the channel B is on CZ, in the first 30 seconds of recording with eyes open, in the next 20 seconds of recording with eyes closed, then 45 seconds of eyes open, and then 45 seconds of cognitive activity (clients were told to count backwards from 1 to 20) and then 15 seconds of eyes open and finally 30 seconds of harmony are played. In the second stage, the channel A electrode is fixed and the channel B is moved to O1, and the first 30 seconds of recording is performed with eyes open, the next 20 seconds of recording is performed with eyes closed, and finally, 20 seconds of recording is performed with eyes closed. In the third stage, channel A is fixed and the channel B is moved to point F3, and 45 seconds of recording is performed with eyes closed. In the fourth stage, channel A is fixed and channel B is moved to point Z F, and 45 seconds of closed-eye recording is performed. The entire process takes 5 minutes and 20 seconds.

Cognitive rehabilitation protocol: This program was implemented in 16 60-minute individual sessions as follows.

LORETA NF training: In LORETA NF, sensors, called electrodes, are placed on the patient’s scalp. These sensors record the electrical activity of the individual’s brain and show it to him in the form of brain waves (often in a simulated form in the form of a computer game or video). In this case, the movie or computer game is played without using hands and only with the person’s brain waves. In this way, by seeing the progress or stop of the game and getting rewards or losing points or changes in the sound or movie playback, the person learns the favorable or unfavorable conditions of his brain waves and tries to correct the state of his brain wave production by guiding the game or movie.[27] In general, in this method, the child sits in front of a monitor, and the electrode placed on the child’s head receives the brain waves and directs them to the computer. By seeing his brain waves on the monitor screen, the child can adjust and control them, and throughout these steps, the therapist sits next to the child and teaches the child how to do this. The students received treatment in 15 30-minute sessions.

Ethical consideration

This randomized, parallel-group clinical trial was conducted over a two-month period, affiliated with Azad University of Shiraz. The trial protocol received approval from the ethical committee of Shiraz University (approval ID: IR.IAU.SHIRAZ.REC.1402.315). Moreover, the trial procedures adhered to the principles outlined in the Declaration of Helsinki.

Statistical analysis method

Descriptive statistics, including mean and standard deviation, were used to analyze the data, and multivariate analysis of covariance was used for inferential statistics.

Results

Fifteen people were placed in the control group, 15 in the cognitive rehabilitation group, and 15 in the NF group. The descriptive statistics of the three groups of study are shown in Table 1.

Table 1.

Descriptive statistics of three groups of study

Variable		Group		Stage		n		M		SD
Working memory		Control		Pretest		15		50/40		8/700
				Posttest		15		51/87		7/615
				Follow-up		15		51/07		7/353
		Cognitive rehabilitation		Pretest		15		51/03		11/121
				Posttest		15		57/20		11/478
				Follow-up		15		54/40		12/432
		NF		Pretest		15		50/80		7/757
				Posttest		15		60/40		8/348
				Follow-up		15		58/20		8/368

NF=Neurofeedback, SD=standard deviation

As can be seen, in the control group, the average scores in the posttest and follow-up stages did not show much change compared to the pretest, but in the experimental groups, we saw an increase in scores in the posttest and follow-up stages compared to the pretest [Table 1]. The Z scores in pretest, posttest, and follow-up stages are shown in Table 2.

Table 2.

Z scores in pretest, posttest, and follow-up stages

		Pre test		Sig		Posttest		Sig		Follow-up		Sig
Variable		z				z				z
Working memory		0/287		0/143		0/354		0/135		0/694		0/102

Based on the results in the table, the statistical significance level calculated for all variables is greater than 0.05, so the assumption of normal distribution of scores is accepted [Table 2]. The significance level of the Box test is 0.095. Since this value is greater than the significance level of 0.05 required to reject the null hypothesis, the null hypothesis of homogeneity of the covariance matrices is accepted.

As can be seen, the significance level of all four relevant multivariate statistics, namely the Pillai effect, Wilks’ lambda, Hotelling effect, and largest square root, is less than 0.01 (P < 0.01) [Table 3]. Thus, the null hypothesis is rejected and it is determined that there is a significant difference between the working memory scores in the cognitive rehabilitation and NF groups at the posttest stage. To examine the differences between the groups in each of these variables, the between-subjects effects test was used, the results of which are presented below. Results of the univariate within-subject effects test for comparing working memory between the three groups are shown in Table 4. Results of the multivariate within-subject effects test for comparing working memory between the three groups in follow-up stage are shown in Table 5.

Table 3.

Results of the multivariate within-subject effects test for comparing working memory between the three groups

		Tests		Amounts		F		Df1		Df2		Sig		Effect size
Effect		Pillai effect		0/468		4/618		4		21		0/008		0/468
		Wilks’ lambda		0/532		4/618		4		21		0/008		0/468
		Hotelling effect		0/880		4/618		4		21		0/008		0/468
		Largest square root		0/880		4/618		4		21		0/008		0/468

Table 4.

Results of the univariate within-subject effects test for comparing working memory between the three groups

Variable		Source		SS		df		MS		F		sig		Effect size
Working memory		Between groups		76/164		1		24		76/164		0/015		0/221
				268/314		1		11/180

Table 5.

Results of the multivariate within-subject effects test for comparing working memory between the three groups in the follow-up stage

Effect		Tests		Amounts		F		Df1		Df2		Sig		Effect size
		Pillai effect		0/551		6/453		4		21		0/551		0/002
		Wilks’ lambda		0/449		6/453		4		21		0/551		0/002
		Hotelling effect		1/229		6/453		4		21		0/551		0/002
		largest square root		1/229		6/453		4		21		0/551		0/002
Variable		source		SS		df		MS		F		sig		Effect size
Working memory		Between groups		120/442		1		120/442		7/392		0/012		0/235
				278/792		24		11/616

The significance level of the Box test is 0.095. Since this value

By comparing the mean scores of the two groups, it is observed that in the NF group, compared to the cognitive rehabilitation group, we see a greater increase in working memory scores in the follow-up phase, indicating a greater continuation of the effectiveness of NF treatment compared to cognitive rehabilitation.

Discussion

The current double-blind clinical trial was designed to evaluate and compare the effects of low-resolution electromagnetic tomography (LORETA) neurofeedback and conventional cognitive rehabilitation on working memory performance in children diagnosed with learning disorders (LD). Our results demonstrate that while both interventions significantly improved working memory, the group receiving LORETA neurofeedback showed significantly greater improvements across several working memory measures. These findings highlight the potential superiority of LORETA neurofeedback in addressing the cognitive deficits that are characteristic of learning disorders.

The superior efficacy of LORETA neurofeedback may be attributable to its ability to directly modulate deep cortical regions associated with working memory and executive function.[2] Previous neuroimaging and electrophysiological studies have identified that children with LD often show dysregulation in the frontal and parietal cortices, particularly the dorsolateral prefrontal cortex (DLPFC), anterior cingulate cortex (ACC), and posterior parietal cortex—regions critical for attention, working memory, and cognitive flexibility.[6] LORETA neurofeedback, by estimating and training electrical activity in these source-localized regions, likely facilitates improved functional connectivity and self-regulation of these networks, leading to more pronounced cognitive gains.[5,15] In contrast, traditional cognitive rehabilitation relies on task-based behavioral training intended to reinforce specific cognitive strategies and improve metacognitive skills through repetition and adaptive task difficulty. Although effective to a degree, this method may not sufficiently engage or normalize the underlying neural circuitry that gives rise to impaired working memory function. Thus, while cognitive rehabilitation remains a valuable therapeutic tool, it may not yield the same depth of neurological change as neuromodulation techniques like LORETA neurofeedback.[17,22]

Importantly, the present findings align with the growing body of research supporting the use of neurofeedback in developmental neuropsychology. Neurofeedback interventions have been shown to improve attention regulation, emotional control, and learning capacity in children with various neurodevelopmental disorders, including ADHD, autism spectrum disorder, and specific learning disorders. LORETA, as an advanced form of neurofeedback, offers an individualized, precise method of targeting abnormal neural activity, potentially resulting in a more efficient and enduring treatment effect. Furthermore, our use of a double-blind design strengthens the internal validity of the study by reducing bias from participants, caregivers, and evaluators. The use of standardized, psychometrically sound working memory assessments ensures the reliability of observed outcomes.[21,27] These methodological strengths provide robust support for the conclusion that LORETA neurofeedback may be a more effective intervention than cognitive rehabilitation for improving working memory in this population.[9] Beyond the cognitive domain, enhanced working memory has far-reaching implications for children with LD.[10] Working memory serves as a critical foundation for academic skills, such as reading comprehension, mathematical reasoning, and written expression. Improvements in working memory may, therefore, translate into better school performance, improved task persistence, and increased confidence in academic settings.[1,11] Additionally, better cognitive functioning could reduce the behavioral and emotional challenges often observed in children with LD, such as frustration, low self-esteem, and inattentiveness.[12]

The clinical implications of these results suggest that integrating LORETA neurofeedback into intervention programs may offer a powerful, non-pharmacological tool to support cognitive development in children with learning disorders. Given its neuroplastic effects and individualized approach, LORETA may be particularly useful in cases where traditional behavioral therapies have produced limited success. It also opens new avenues for personalized medicine in educational and clinical neuropsychology, offering interventions tailored to each child’s unique neural profile.

Limitation and recommendation

Although the study employed a randomized, double-blind design, the sample size may have limited the statistical power to detect small-to-moderate effects or interactions between subgroups (e.g., age ranges, severity of learning disorders). The study assessed outcomes shortly after the intervention. Long-term effects of LORETA neurofeedback and cognitive rehabilitation on working memory were not evaluated, limiting conclusions about the durability of treatment gains. Moreover, the primary outcome was working memory, which is only one aspect of cognitive function. Broader executive functioning and academic performance were not assessed, potentially overlooking important treatment effects or trade-offs. So, future studies should include larger and more demographically diverse samples to increase external validity and support subgroup analyses (e.g., by age, type of learning disorder). Research should explore the potential benefits of combining LORETA neurofeedback with cognitive rehabilitation or other evidence-based interventions.

Conclusion

In conclusion, while LORETA NF demonstrates promising advantages over cognitive rehabilitation for improving working memory in children with learning disorders, careful consideration of implementation challenges and individual factors is essential. Future research should focus on addressing current limitations while developing more accessible and standardized protocols. The potential for integrating elements of both approaches may ultimately provide the most comprehensive and effective intervention strategy for this population.

Informed consent

All participants provided their written informed consent to participate in this study.

AI policy statement

The authors confirm that no generative AI or AI-assisted technologies were used in the conception, data analysis, or writing of this manuscript.

Ethical consideration

The ethics committee of Azad University of Shiraz Branch approved all study protocols (IR.IAU.SHIRAZ.REC.1402.315). We conducted the research adhering to the principles of the Declaration of Helsinki.

Author’s contributions

SA contributed to the conception and design of the study. JK contributed to the and preparation of materials, data collection, and analysis. The manuscript was written collaboratively, and all authors have read and approved the final version.

Data reproducibility

All data are available and presented by formal request from corresponding author.

Conflict of interests

The authors declare that they have no competing interests.

Funding Statement

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.