Analyzing EEG waves to support the design of serious games for cognitive training

Abstract

In this paper, electroencephalography (EEG) waves were analysed during the use of a developed video game. It implements specific mechanics (stimuli) that promote the training of cognitive abilities such as short-term memory, attention, and concentration. The conducted study involved ten participants: healthy children between 7 and 12 years old (developmental age) who played the game for an average of 19.2 min. The research question for the present work refers to how specific mechanics involved in platform video games activate electroencephalogram waves according to the above-mentioned cognitive processes. In this sense, the responses have been analysed in terms of cognitive activation through EEG during the video game mechanics. The results are consistent by showing significant findings, described in the manuscript conclusions, and related to EEG wave magnitudes during the studied game mechanics. These findings contribute to supporting the design of video games with cognitive training goals.

Appendices

Appendix 1: Game elements

Game elements are techniques that we use to motivate and engage the user and influence people’s behaviours, in our case to promote the intensive use of the video game for potentially training cognitive skills. These mechanics involve the achievement of a series of objectives that can be translated into a reward. To adapt the different game mechanics, it is necessary to identify the goal of the game, to create new challenges that develop different abilities, and to define the reinforcement or motivation.

Following the taxonomy of gamification elements (Hervás, 2017), there are six main game element categories (in bold), five of which are included into the Api’s Adventures video game. For each category, several specific techniques were implemented (in italics):

• Immersion This corresponds to the deep mental involvement in the game context. Two techniques of this category were included: the player assumes the role of a young Martian (Api), shown in Fig. 12, in a narrative story where Api must find the path to get back home and faces several obstacles along the way. It is very important in video games to follow a storyline because it is related to attention, memory or concentration.

• Goals The game is divided into levels (10 stages, as shown in Fig. 13) following the information cascade theory to provide an increasing way to understand the game and guide the user’s learning. The goals of the game follow the Csikszentmihalyi (1990) rules: it must be a challenge must it not be overly complicated; goals must be designed clearly and contribute to the “paradox of control”, i.e. the user understands that the result of the activity is uncertain but must have the feeling that the end result depends on his or her actions. The goals require the interruption of automatic responses and monitoring of the performance, according to the task and the adequate strategy (concentration). Consequently, the player would need to plan the correct sequence of actions to achieve the goal and to actively maintain this set of actions (Montani et al. 2014) related to short-term memory.

• Scoring Each stage includes the evaluation through points of how well the user completed each level, using the metaphor of collectables (“diamonds”). Also, the user’s performance is determined by the number of meters he or she has been able to move forward. All these elements are shown in Fig. 14, and they contribute to providing feedback: users aim to identify their performance and defeats, and the information provided will serve to identify possible improvements in their strategies for playing the game.

• Randomness The infinite-game mode provides permanent variations of the game due to the procedural implementation, where each round is totally different due to the run-time generation.

• Status A ranking of best players has been included in the infinite game mode, based on how long the user kept Api alive.

Fig. 12

Player assets for the main character, Api

Fig. 13

Level-based organization with incremental difficulty and progressive learning

Fig. 14

Scoring implemented through numbers of meters and collected diamonds

Appendix 2: Engagement and motivation

During the performed study, the usability of the developed game was evaluated. The evaluation of usability has been made to ensure that the video game has been properly designed regarding user motivation and engagement and, particularly, to assess whether the user feels comfortable playing the game, its ease of use and the coherence of the included game elements. To evaluate usability, we used a specific questionnaire called the System Usability Scale (SUS) (Brooke 1996). This questionnaire consists of 10 items, with odd-numbered items worded positively and even-numbered items worded negatively. All participants completed a version of the SUS questionnaire translated to Spanish. The translation was made by the authors of this paper, with efforts made to being faithful to the original version. The translated version was not assessed in terms of validity or reliability, but the authors consider that this constraint is not a critical limitation for the goal of this work.

Table 2 shows the results of the SUS questionnaire per user and per items, providing some statistical measures. Once the results shown in Table 1 were analysed, we were able to conclude the following: first, it is necessary to note that 80% of our participants were frequent video-game players. The two participants who did not play video games frequently had the lowest scores in the SUS questionnaire (participants S5 and S7). The video game was inspired by classic platform video games, and users who had played similar games felt more comfortable playing Api’s Adventures.

Table 2 Results of the SUS questionnaire per user (columns) and per item (rows) including mean ($\stackrel{-}{\rm{x}})$, median ($\stackrel{~}{\rm{x}}),$ standard deviation $\left({\upsigma }\right),$ and mode $\widehat{(\rm{x}})$

The global average score was 76.75. Based on industry standards (Sauro 2011), the score to get a letter-grade of A (on a scale from A + to F) should be above 80.3, so usability can be improved. The obtained averaged score corresponds to a mark of B. Focusing on particular items, those that deal with motivation and engagement obtained better results, with an average score of 82.5 (A). The main aspects needing improvement are related to the ease of understanding how to play the video game, an average of 75 points (B), and confidence regarding the video game, which was 72.5 (B).