| Description | As with all software, the quality of games designed for serious applications – such as education, training and health – is extremely important. The quality characteristics expected of a piece of software are typically viewed as non-functional requirements, such as reliability, usability, and so on. These are distinguished from what are usually known as the software’s functional requirements, that is, the features the software must incorporate (e.g. a word processor should allow the user to delete text). Several standards have been developed to characterize software quality, such as the International Standards Organization’s ISO/IEC 25010, which defines quality in use and product quality models for general software. Games, however, present unique challenges in the assessment and assurance of software quality. Unlike other software artefacts such as word processors, successful games must invoke a wide variety of emotional experiences in the player. Serious games present yet a further challenge: in addition to quality requirements such as fun, immersion, and so on, serious games must also (by their very definition) achieve a measurable outcome. Whether it is effectively training the cognitive process they are designed to train, or effectively promoting the health behavior they are designed to promote, the quality of a serious game must be partly evaluated in terms of the non-entertainment purpose that makes it a serious game. Crucially, whether or not a serious game is effective in this sense is not solely dependent on the kinds of non-functional quality characteristics addressed by quality assurance models. It is also important that the gameplay of a serious game effectively promotes the desired behaviors, skills, or knowledge. These gameplay features are best understood as the functional requirements of the serious game. This suggests that the relationship between gameplay features (functional requirements) and software quality (non-functional requirements) is both important and complex in serious games. A novel model of this relationship is presented here. The model is informed by: (1) consideration of the relationship between the functional requirements and quality characteristics of games in general; (2) analysis of a recently-proposed quality model for serious games, based on ISO/IEC 25010; and (3) practical experience in the development and evaluation of serious games. Implications of the model are discussed: it is argued that quality evaluation should occur early in the development of serious games, and that the development of serious games can particularly benefit from iterative, agile approaches. |
