The brain is a very complex network of nonlinear systems with internal dynamic states that are not easily measurable. The networking nature of the brain is very much related to different functional tasks, such as memory, which is built up in scattered areas of the brain [1].

There are several classifications for memory, including sensory, short-term, and long-term memory. In recent decades, the term of working memory has replaced short-term memory. Working memory includes the ability to store and manipulate information for a short term and can be divided into subcategories of primary encoding, information, and storage and retrieval of materials [2]. People have limited capacity for information that can be stored in the working memory in a given time. Neuroanatomically, the working memory network is distributed over a large part of the brain. In particular, the task of working memory involved in visuospatial information activates the prefrontal cortex [3]. In addition, prefrontal and parietal neuronal accumulation are activated simultaneously during the maintenance of working memory [4]. Increased difficulty in a working memory task is associated with increased communication between the frontal and parietal areas [5].

Cortical activity can be recorded non-invasively on the scalp using electroencephalography (EEG). Since the EEG is nonlinear and unstable, phase synchronization has been well studied to determine the cortical networks. Phase synchronization is a statistical method of measuring the interdependence of two oscillators and it is used in the field of nonlinear dynamics and irregular systems [6]. Synchronization of the short-term or localized phase in the brain can be interpreted as the cause of the regional perception link [7]. Synchronization of the long-term phase between areas includes motor planning [8], excitement [9], and memory [10]. 

The results of previous research have shown that an increase in the synchronization of the electrocortical stage across the cortex facilitates neuronal connections, enhances neuronal morphology, and supports working memory [10]. The synchronized firing of pre-synapse neurons increases the rate of post-synapse neuron firing [11]. Synchronized gamma oscillations are limited to the local neuronal region, while theta synchronization is more effective over long distances (e.g., different areas of the brain). Successful data encryption during a work memory task requires increasing step synchronization [12]. Phase synchronization in the theta band is maintained between the prefrontal and parietal areas during a working memory task over encryption, maintenance, and retrieval [13]. Beta and gamma synchronizations, due to excitation, increase the coherence between the forehead and parietal areas while maintaining working memory [14]. Synchronization can be non-invasive by providing a special sound stimulus called binaural beats. Two-way beats require two different soundtracks to be heard [15]. This method causes a beat of the two-third phones, the frequency of which is equal to the difference between the two tones provided, which is produced in the Inferior Colliculus (IC) located in the auditory canal [16]. General phase difference from IC to auditory cortex with periodic nerve firing is maintained at binaural beats frequencies [17]. The highest synchronization rate occurs due to the binaural beats in the auditory cortex on the beta band at 16 Hz [18]. 

Based on the findings of previous studies, binaural beats affect the cortical responses throughout the frequency bands. In the gamma band, the largest EEG steady-state responses occur at a rate of 40 Hz and it mainly activates the frontal and parietal lobes. In addition, the stimulation of binaural beats in the beta band on 5.18 Hz, increases EEG size by 21% [19]. The cortical areas are subjected to the binaural beats theta bands, including the parietal, frontal, and temporal regions [20].

The results of the research have shown that the binaural beats affect many aspects of cognition and mood, such as attention, memory, creativity, and alertness [21]. Although this method offers a non-invasive and comfortable stimulus, its effect on working memory has only been studied in a limited number of studies. For example, 22. Beauchene et al. [22] indicated that the binaural beats of 15 Hz could strengthen the accuracy and key connections in cortical networks in a task working memory. Klimesch et al. [23] showed that fluctuations in the alpha rhythm indirectly increased the performance of working memory capacity. So that they filter out inappropriate information and prevent the disruption of processes in the form of contradictory stimuli. On the other hand, Beauchene et al. [24] showed that the binaural beats on the beta band at 15 Hz during the execution of the working memory task activated significant frequency bands and created networks that were characterized by the transmission of more information. Carter and Russell [25] exposed 8-12-year-old boys with learning disabilities to 10 Hz and 18 Hz binaural beats stimulation sessions for 8 weeks and found an improvement in Raven’s progressive matrices and in a subtest of auditory sequential memory. McMurray [26] evaluated the multiplication of Duchess 7 and 11 Hz on Alpha's brain wave, working memory, and the attention of healthy elderly people. Exposure to double beeping for 2 min causes a change in the electrical activity of the brain. In particular, changes in alpha wave activity occurred. In addition, there have been advances in the direct and indirect digit span tasks and in a part of the continuous performance test. Dai et al. [27] also showed that there was a higher functional integration in the theta band in the task of working memory than in the control group.

According to the above, the binaural beats can induce neuronal fluctuations and affect various cognitive functions, including working memory. On the other hand, research on the use of different frequencies to improve working memory shows conflicting results. Moreover, a more appropriate and specific band is not specified to enhance working memory.

Objectives

Therefore, this study aimed to investigate the differences between the effects of stimulating the alpha, beta, and theta bands by binaural beats on students' working memory?

Materials and Methods

The present quasi-experimental study was conducted based on a pretest-posttest control group design. The statistical population of the present study included all students aged 19 to 25 years at Azerbaijan Shahid Madani University, Tabriz, East Azerbaijan Province, Iran, studying for a bachelor's degree in the 2018-19 academic year. The samples (n=60) were selected using an available sampling method and then randomly assigned to three experimental groups and a control group. The exclusion criteria from the study were taking psychiatric medications and having a major depressive disorder, anxiety disorders, a history of brain trauma, a history of epilepsy or suffering from it, and heart disease. The collected data were analyzed in SPSS software (version 18) using a one-way analysis of covariance.

Research tools

The Corsi block tapping test was developed and used in the early 1970s [28]. This test is designed based on the Digit Span test; however, this test requires the use of visuospatial working memory rather than the verbal form available in the Digit Span test. Functional magnetic resonance imaging studies on the tests that were being performed on the Corsi test have indicated that an increase in the number of sequences and stimuli of this test does not change the overall activity of your brain [29]. Consequently, while people may have trouble encoding, this increase in test levels has no relationship with general brain activation. In the general case, in the direct Span of Corsi test, we need the support of a visuospatial sketchpad and there is no need for a phonological ring, and when the sequence of recovery cases is more than three or four items, the central executive resources are used [30].

Within the process of the Corsi test, the subject sees nine blocks on the computer screen that in each attempt, several of these blocks are illuminated with a specific sequence. The subject's task is to remember the sequence of blocks being lit. Afterward, the subject repeats the sequence by clicking on the blocks. This test starts with two blocks and gradually increases the number of bright blocks in each attempt until it reaches 9 blocks. If the subject makes a mistake twice in a sequence, the test ends, and the longest sequence reminded by the subject is recorded. In general, the average reminder for normal people is five blocks [31]. The reliability of this test has been reported to be 0.73 [32].

Binaural beats stimulation

The Binaural beat frequency utilized in this study is equal to the difference between the frequencies used. In this regard, the following frequencies were produced by Audacity software (version 2.2) and Adobe Audition CC (version 2017) and high-quality wireless headphones: 1) for the right ear = 220/5 Hz and for the left ear = 230 Hz (9.5 Hz - alpha range) according to the protocol used by Kraus and Probanova [33], 2) for the right ear = 240 Hz and for the left ear = 255 Hz (15 Hz - beta range) according to the protocol used by Biochan et al. [22], and 3) for the left ear = 250 Hz and for the right ear = 256 Hz (6 Hz - theta range) according to the protocol used by Giracitiaacorn and Wangsawat [34]. 

Research method

Results

The present study involved 60 students, and all subjects in four groups had the same education level and were undergraduate students. All participants in the pre-test and post-test stages were evaluated in terms of visuospatial working memory by the Corsi block tapping test. One-way analysis of covariance (ANCOVA) was used to perform statistical analysis. Before the ANCOVA test, its assumptions were examined. The results of the Kolmogorov-Smirnov test showed that the data had a normal distribution (sig=0.11, z=1.19). Based on the results of Levene's test, the assumption of homogeneity of variances was observed (sig=0.074, f=0.001).

Descriptive findings indicated an increase in the working memory scores of the experimental group in the post-test stage.

As shown in Table 2, after deleting the pre-test effect, there was a significant difference in the post-test scores between the experimental group and the control group (P<0.05). Accordingly, synchronizing brainwaves improved visuospatial working memory and enhanced working memory capacity. Moreover, the square of Eta was obtained at 0.14.

As the Bonferroni test shows, there was a significant difference between the brain wave synchronization groups in the beta band and the control group (P=0.30). However, there was no significant difference between alpha and 

beta, beta and theta, alpha and theta, alpha and control groups, and theta and beta (P<0.05).

Table 1. Mean and SD of visuospatial working memory per group

*Significant difference between groups (P<0.05)