The Psychological Impact of Colour and Music on Concentration

. This paper details the research conducted as vacation work undertaken for Touch Prosthetics during which an electroencephalogram (EEG) headset was used to analyse psychological priming effects on subjects’ concentration levels using visual and audial stimuli. The experimental procedure consists of a visual aspect and an audial aspect during which a participant’s brain activity was measured when exposed to their self-reported favourite colour or music genre, respectively. The findings of both aspects of the experiment may have applications in an industrial environment as a measure to improve employee concentration, and consequently ensure health and safety by minimising hazards due to poor concentration.


Introduction
The health and safety, productivity, and concentration levels of employees are directly influenced by the physical aspects of the environment in which they work [1]. Optimal concentration levels are particularly important in industrial environments since employees work with potentially hazardous machinery, tools, and chemicals. Therefore, any mistake or error as a result of poor concentration could result in serious health and safety threats such as harm, injury, or death to those responsible and others. The following report demonstrates the effectiveness of neuroscience in relation to the aforementioned hazards by illustrating how EEG analysis can be used to improve concentration levels through priming effects with either a participant's self-reported favourite music genre or colour.
Priming is a psychological phenomenon whereby a subject's performance during a cognitive task, such as concentration, can be improved by exposure to some stimulus, without conscious intention [2], [3]. Priming effects using audial and visual stimuli have been proven to affect a subject's behaviour [1], [4]. Psychological priming may be used to improve concentration in industrial working environments through the implementation of visual stimuli such as colourful cafeterias, workspaces, and toilet areas. Employees should then be encouraged to utilise the facilities that correspond to their self-reported favourite choice of colour. Priming may also be used through the implementation of audial stimuli by encouraging employees to listen to their self-reported favourite music genres before work and during lunch breaks, in an effort to improve concentration throughout the workday.
EEG is an electrophysiological process that measures change in the electrical activity produced by the cerebral cortex of the brain [5]. The cerebral cortex is the outermost layer of the brain and is divided into the left hemisphere and right hemisphere. Each hemisphere is further divided into four lobes, i.e., frontal, temporal, parietal and occipital.
There are various non-invasive EEG devices that are commercially available and have a range of different features to suit the needs of consumers and researchers. These devices are used to measure brain waves and frequency bands by placing EEG electrodes along a participant's scalp. There are four main frequency bands that describe EEG signals: Delta (0.5-4 Hz), Theta (4-8 Hz), Alpha (8)(9)(10)(11)(12)(13)(14)(15) and Beta (15-30 Hz) [5].
The EEG headset that was used to conduct the experiments as detailed in this report was the EPOC + Headset. The EPOC+ records high-resolution EEG data and provides coverage of both hemispheres of the brain [5]. The EPOC+ headset was also the most credible and costeffective EEG headset when compared to other EEG devices offered by Emotiv Inc. such as the EPOC Insight headset and the EPOC Flex cap [5]. The software that was used to conduct the experiments in this report was Testbench which is a Software Development Kit (SDK) from Emotiv Inc.
Most EEG headsets, including the EPOC+, are subject to limitations such as data noise interference caused by participants' muscle movements, eye blinks and environmental factors [6]. Another limitation was that electrode contact quality could not be reached at 100% during the experimental procedure. Therefore, measures were put in place during the experiment to ensure that the test results were accurate such as taking ten sets of results for each experiment and designing the procedure such that no movement of subjects were required during testing.170 x 250 mm paper size (W x H mm) and adjust the margins to those shown in the Table 1. The final printed area will be 130 x 210 mm. Do not add any page numbers.

Reparation of equipment
The equipment accompanying the EEG headset consisted of the 16 sponge pads, 16 electrodes and 2 reference electrodes. Since the headset was aged, subject to wear and tear, and unused for a long time, there were several damages to the equipment. These damages consisted of electrode oxidation, broken twist and lock mechanisms of all the electrode holders, and four electrode holders that were broken in half.
Since the damage to the equipment was extensive, the following steps were undertaken to repair the electrodes prior to the first use of the apparatus: 1) The battery of the headset was dead and had to be charged for 24 hours prior to the first use.
2) The sponge pads were hard and filled with green crystal formations caused by electrode oxidation. The sponge pads were repaired by soaking them overnight in a container filled with 0.9% concentrated saline solution to soften the sponge pad and remove the crystal formation.
3) The electrodes also had green crystal formations that were caused by electrode oxidation. This was cleaned using cotton ear buds and a diluted solution of water and isopropyl alcohol. Caution was taken not to damage the coating of the electrode.

4)
The twist and lock mechanism of all the electrode holders were broken so that they could not be secured to the headset. This issue was fixed by cutting approximately 70mm×10mm strips of black masking tape and using these strips to secure the electrode holders to the headset.

5)
It was also found that four of the electrode holders were broken in half. This issue was rectified by first rolling pieces of reusable rubber-like adhesive into balls of approximately 0.5mm diameter. These balls were then placed on either end of each broken piece. The corresponding broken pieces of the electrodes were then joined and sealed until they formed a closed ring. Any excess adhesive protruding from the sides of the electrode holders were removed using a tweezer.

Preparation and setup of equipment
The following steps were mandatory before conducting each experimental procedure: 1) The headset was charged for a maximum of four hours until the battery was at full capacity.
2) The sponge pads were rinsed in water and the excess liquid was drained using a paper towel.
3) The electrodes were carefully cleaned using a cotton earbud and saline solution such that the electrode coating was not damaged.

4)
The electrodes were placed in their corresponding holders. Steps 4 and 5 as mentioned in section 2.A were then repeated.

5)
Once all the electrode holders were secured to the headset, the sponge pads were placed in each holder. A few drops of saline solution were then applied to each sponge pad.
6) The two reference electrodes were fastened onto the headset so that they would rest behind the subject's ears.

7)
The dongle of headset was connected to the laptop and the relevant software was opened.

8)
The headset was switched on and fitted onto the subject's head.
9) The electrodes were adjusted and worked through the subject's hair so that good contact quality was made between the sponge pads and the subject's scalp.

Experimental Procedure
The experiment consisted of a visual aspect, and an audial aspect. The visual experiment was conducted on Subject 1, a female of 50 years of age with no record of eye issues or neurological disease. The audial experiment was conducted on Subject 2, a female of 24 years of age with no record of hearing issues or neurological disease.

Visual Experiment
During the visual aspect of the experiment, four different colour boards, i.e., red, yellow, green, and blue, were used during testing. A white board served as the control for the experiment. A screen was set up on a blank wall with the colour boards setup in the order described in Table 2. The procedure of the experiment was discussed with Subject 1 beforehand so that no movement and talking occurred during the actual experiment. A presurvey was also conducted in which the subject stated that their favourite colour was blue. The experiment was conducted in the following order: 1) The equipment was prepared, and setup as described in steps 1 through 9 of section 2.B.
2) The subject was instructed to sit on a chair facing the wall with the colour screen as described in Table 1.
3) The new .edf document was created on the Testbench Software and labelled as the subject's name and test recording number, e.g., Subject1Test1.
4) The subject was instructed to begin the experiment by looking at the screen.
5) The subject viewed each colour board for 10 seconds in the corresponding order as mentioned in Table 1.
6) The recording was stopped and saved after the final time interval.

7)
Each .edf file was converted into a .csv file via Testbench so that the raw EEG data could be analysed.

8)
The headset was removed from the subject's head, marking the end of that particular test.

9)
Steps 1 through 8 were repeated until 10 sets of test results were obtained.

Audial Experiment
A recording of four different music genres, i.e., classical, rock, hip-hop and pop, as described by Table 2 was created. The silent interval served as the control for the experiment. Each of the four chosen songs in the recording are 120 beats per minute (BPM) which referred to a measure of the music tempo. This was done to eliminate music tempo as a variable since music tempo is known to affect beta wave activity [7]. The procedure of the experiment was discussed with the subject beforehand so that no movement and talking occurred during the actual experiment. A pre-survey was also conducted in which the subject stated that their favourite genre was hip-hop. The experiment was conducted in the following order: 1) The equipment was prepared, and setup as described as detailed in steps 1 through 9 of section 2.B.
2) The subject was instructed to sit comfortably on a chair in a quiet room 3) The new .edf file was created on Testbench software and labelled as the subject's name and recording number, e.g., Subject2Test1.
4) The subject was instructed to begin the experiment by closing their eyes so that visual distractions were minimised.
5) The recording as specified in Table 2 was played.
6) The recording was stopped, and the file was saved after the final time interval.
7) The test recording was converted into a .csv file via Testbench so that the raw EEG data could be analysed.
8) The headset was removed from the subject's head marking the end of that particular test. 9) Steps 1 through 8 were repeated until 10 sets of results were obtained.

Test Results
Test results were obtained from the raw EEG data displayed on Microsoft Excel sheets and from the actual recordings in which a fluctuating bar graph illustrated the signal power of each frequency band, Alpha, Delta, Beta and Theta, was paused at each required time interval. The type of brainwave frequency that was chosen to be analysed was beta (13 -30Hz) since beta waves have been shown to increase during concentration in the frontal and occipital lobes [8]. Beta waves have a frequency of 13-30Hz and are responsible for a subject's conscious processing of information, active attention, and problem solving [9], [10]. Beta waves also represent a state of alertness and concentration which keeps an individual sharp and focused [11].

Electrode Analysis
The EPOC + headset has 14 electrodes channels, i.e., AF3, F7, F3, FC5, T7, P7, O1, O2, P8, T8, FC6, F4, F8, and AF4, that are placed around a subject's head. Even though all the electrodes were connected to the headset during the experiments and test results were obtained, only certain electrodes were chosen for analysis so that uniformity was maintained. These electrodes were chosen according to the following reasoning: • The electrodes that showed the most activity when analysing the raw EEG data of the 10 data points of the visual experiment were: AF3, AF4, F8, F7, F3 and FC5. Therefore, the other electrodes T7, P7, O1, O2, P8, T8, FC6, F4 were eliminated as possible options for the visual experiment. Most of the same electrodes, AF3, AF4, F8, F7 and FC5, showed the most activity for the audial experiment. However, there was significant activity for electrode O2 during the audial experiment. • The electrodes F7, F8, FC5 and F3 did not maintain contact quality for all experiments. Therefore, these electrodes were eliminated as possible options since 370, 05003 (2022) https://doi.org/10.1051/matecconf/202237005003 MATEC Web of Conferences 2022 RAPDASA-RobMech-PRASA-CoSAAMI Conference 10 complete sets of data could not be obtained for both the visual and the audial experiment.
• Therefore, the only remaining pair of electrodes, AF3 and AF4, were chosen to be analysed for both the audial and the visual experiment. However, since there was a significant amount of brain activity for O2 during the audial experiment, this electrode was also chosen to be analysed.

Test results for the visual experiments
The test results of the visual experiment were taken at the middle values of the specific time intervals for each colour as shown in Table 1. This was done to ensure that there was no interference or false readings from the movement of pages when the colour board was changed. From the 10 data points taken from each electrode for the four music genres tested, the mean signal power values of beta frequencies were calculated using the formula as shown in (1) and were summarised as shown in Table 3. Where: ̅ : Mean (dB) ∑ : The summation of data points : The number of data points The data shown in Table 3 was used to generate the graph as shown in Fig. 1 which illustrates the relationship between colours and the mean signal power of beta frequencies for electrodes AF3 and AF4.

Test results for the audial experiments
The test results of the audial experiment were taken at the middle values of specific time intervals for each music genre as shown in Table 2. This was done to ensure that there was no interference or false readings from the initial shock caused by the change of the music genre. From the 10 data points taken from each electrode for the four music genres tested, the mean signal power values of beta frequencies were calculated using (1) and were summarised as shown in Table 4. The data shown in Table 4 was used to generate the graph as shown in Fig. 2 which illustrates the relationship between music genres and the mean signal power of beta frequencies for electrodes AF3, AF4 and O2.

Conclusion
For the visual experiment, there was a slight correlation between the subject's self-reported favourite colour, i.e., blue, and improved concentration levels for electrode AF3. However, as shown in Fig.5, there was no correlation between concentration levels and the colour blue since the greatest beta activity was observed when the participant viewed red. The audial experiment was more successful than the visual experiment since as illustrated in Fig. 6, the mean signal power for beta frequency were significantly higher for the subject's self-reported favourite music genre, hip hop. This demonstrates a direct correlation between a subject's self-reported favourite music genre and improved concentration levels. Hence, the experiment proves that listening to favourite genres of music has shown to improve concentration. The accuracy of future experiments could be improved by conducting research on a greater number of participants. A single participant was chosen for each aspect of the experiment due to time constraints, since a lengthy period of time was required to set up the apparatus, accurately place the electrodes on the subject's scalp, conduct the procedure and to analyse the raw and processed data. Even though necessary precautions and measures were put in place to improve the accuracy and validity of the experiments, there were several errors present, such as noise interference, low contact quality of the electrodes and some degree of distraction of subjects that could have influenced the test results. Therefore, future experimental results could also be improved upon by using newer and superior equipment.