Concepts in HCI

The topics and theories below give insight into the study of HCI. It's critical in HCI to develop a process that better explores user. Using different methods and understandings as tools to investigate and gain insight into user behavior can assist in this. There are many things that affect the user's mental model and behavior such as the environment, past experiences, past learning's, as well as other common situations and tools.

Foundations of HCI

Cognitive Psychology

When developing user interfaces it is important to understand how people gain and use knowledge. This study involves a wide range of areas including human memory, attention, perception, knowledge, reasoning, representation and problem solving. By understanding more about these topics and how they pertain to HCI can help HCI professionals better predict and explain user actions.

1. Information Processing Model: This model was developed by Lindsay & Norman in 1977. This processing model explains the movement of information from input to output and characterizes humans as information processors much like computers. In many ways computers and humans are alike; we encode information, compare, choose appropriate responses, and communicate. By characterizing human behaviors such as these we can better predict and explain user actions.
2. Hacker's Action Theory: Most human actions are driven by goals. Hacker's Action Theory has helped realize that humans tend to maximize efficiency when completing these goals. For example, if someone's goal was to wash dishes they might first stack and categorize all dishes before hand instead of gathering the dishes one by one and washing them individually. Using this theory can help discover patterns in user interaction. This could help a designer with the organization of an interface or menu design depending on the goals the user was trying to achieve. Because of this human characteristic it's vital that the user have the necessary tools/methods to successfully achieve a specific task.
3. Andersen's ACT-R: (Adaptive character of thought rational) Andersen's ACT-R is important for understanding how humans organize knowledge and then convert that knowledge into intelligent behavior. ACT-R helps explain human cognition.
4. Knowledge and Mental Models: It is important to understand how humans gather information and transform that information into knowledge that forms a unique mental model. This assists HCI professionals in realizing how knowledge is represented and organized. This information can be significant in the design and development of applications and tools. The mental model is unique to each individual and it determines how people, think and act. People form these mental models over time through experiences, training and instruction. People tend to solve problems using their mental models using prior information and related experiences. For example, when trying to cool a house one person may set the AC to an extremely cold temperature say 55 degrees to cool a room down as quickly as possible while another person may turn it down just slightly. These two people have different mental models on how this system may work.
5. Attention & Memory: Research on the human memory has helped understand that there are limits on human information processing and memory storage. Knowing this helps HCI experts limit the amount of information that a system should provide. For example, when creating a page that would show a user his bank account details one might think that it would be good to show every piece of information on one screen to reduce the amount of actions taken to locate specific information. However, the amount of information on this page might be overwhelming and hard for the user to process. In this case these tradeoffs need to be recognized.

Perceptual Psychology

As we move through our environment we are constantly obtaining information and turning it into knowledge that is then stored for later use. The study of Perceptual Psychology helps HCI professionals understand how people gather information from graphical representation and interfaces. From the studies in Perceptual Psychology we have learned that people automatically try to make sense and organize information. For example, when humans view a block of information they will mentally arrange it in a way that makes sense to them.

Social Psychology

Social psychology involves the studies of how people act based on the circumstance they are in with others. This is important to HCI because it analyzes the significance of social interactions when people are trying to complete a task. These studies can be of importance when developing message boards or other applications when users interact with each other in groups.

HCI Research

Design Theories

Design theories can assist the design of interfaces and the selection of elements contained within interfaces.

1. GOMS (goals, operators, methods, selection rules) When users are trying to achieve task they create goals and sub goals and achieve those goals using methods and procedures through operators (such as input devices like a mouse or trackball). From these methods users will select a specific method and operators to achieve this task. As stated in Charm Role of HCI Theory document, "the keystroke level predictive model was developed from the above model. This model predicts the performance time of a task by calculating the sum of lesser parts, which include time for keystrokes, pointing, thinking, waiting etc." (http://www.otal.umd.edu/hci-rm/theory.html). It seems as if this could be associated with Hackers Action Theory which states that humans tend to maximize efficiency when completing goals
2. Menu Design: Menu design includes the organizational and selection of information, structure and are an important element of interface development. Utilizing the correct menus can make user performance more efficient and greatly enhance the user's experience. There are many different styles of menus and many trade offs within these styles to consider such as, safety, efficiency and ease of use. When selecting a menu system one might want to take into account the type of application they are developing, the amount of data that needs to be displayed and what the user goals are of the application (efficient process selection, safety, ease of use, etc). Other topics that could be important when conducting menu design include attention and memory.
3. Input Devices: An input device can be described as any device that feeds data into a computer. These types of devices can include keyboards, mouse, trackballs, etc. These devices have their own distinct characteristics and tradeoffs such as speed, ease of use and accuracy. Fitt's law might be considered when determining which input device might be appropriate for a specific task. Fitt's law is an important tool for predicting pointing time, which is the amount of time to move from one distance to another target of a specific width. This law predicts that pointing time increases as a target is further away and smaller targets take longer to point to.
4. User Control/ Direct Manipulation: This states that the user should be in control over the technology and not visa versa. According to Charm Role of HCI Theory document, "The central ideas of user control include: Visibility of object and actions - Rapid, reversible, incremental actions - Replacement of complex language syntax with direct, visual manipulation of the object of interest."( http://www.otal.umd.edu/hci-rm/theory.html) Giving the user more control allows for the user to develop more efficient methods for completing tasks, greater forgiveness when errors are created and allows users to develop there own creative solution for completing tasks.
5. Anthropomorphic Design: Anthropomorphic Design is the design in which computers are programmed to behave like humans. As stated in Charm Role of HCI Theory "Although many people think that this is a good idea for user friendly creations there are some pitfalls. As stated in Anthropomorphism projects false models of the extent of a computer's usefulness and doesn't clarify that there are differences between computers and humans and can injure people emotionally making them feel dumb or inadequate." (http://www.otal.umd.edu/hci-rm/theory.html) This could lead to problems because not all users share a common mental model of how a certain task might be performed. This could be very frustrating to a user when trying to perform a task using a mental model that is different than the one used to program the application.
6. Multimodality: Multimodality is about efficiently utilizing the senses of humans. This combines different methods of communication with man and machine. For example, to enter information into an interface a user could use speech or gesture. This involves investigation into human data processing, perceptual psychology, and working memory processes. This use of human senses provides many uses beyond that of the normal user by extending benefits to people with disabilities. Multimodality could be considered when thinking of Universal Usability, which deals with providing universal access to technology and the tools for people with impairments or disabilities to use these technologies.

Additional Domains of HCI Research

1. Information Visualization: This study examines how to successfully present information visually. This type of work can be seen in creating innovative graphical displays for complicated data. Information Visualization is about displaying information effectively. There are many different techniques of information visualization that help interface designers present data to users in an effective way. Some these techniques include, hiding, layering, proximity, alignment, etc. When thinking about the above one might also want to include attention and memory for the understanding of how much information should be on a page.
2. Ubiquitous & Pervasive Computing: This is about the omni presence of invisible technology in our environments. In this type of environment computers and technology would be used everywhere around us. This can already be seen with the use of Cell Phones, PDA's and Blackberry's all interconnected and constantly surrounding us. In this type of use there would be one person using many computing devices. The only way for this type of world to exist is for the human user interfaces of these devices to be properly designed.
3. Universal Usability: Universal Usability deals with providing universal access to technology and the tools for people with impairments or disabilities to use these technologies. Because of this the designer must understand the types of different users especially ones with disabilities and how they might use a specific system. Government regulations such as section 508 guidelines are assisting in the Universal Usability movement. One might also want to consider multimodality when doing research on this subject.
4. Artificial Intelligence (AI): Can be described as the science and development of making intelligent machines and computer systems. AI developed by MIT seeks to define and understand intelligence. This includes creating computer systems that can mimic human thought; understand speech and other intelligent human actions.
5. Supervisory Control/Autonomous Agents: Autonomous agents are predictive technologies, which are applications that make decisions for users based on the user's habits and behavior. It seems as if Autonomous agents could be used by the department of national security to predict people's behavior. This could be a very complex task with serious drawbacks. Some related topics on this subject might include Anthropomorphic Design and User Control and Direct Manipulation.

Sociology and Social Psychological Theories

1. Social Networks: Are the relationships and flows between groups, people, organizations, etc. This allows HCI professionals to see how social networks including people, animals, etc are mapped together. This network also shows how people are connected together. For instance some people may be connected by work and others by friendship or casual acquaintance. Social networks should also be considered when designing an application that is created for the purpose of bringing people together. One such application is classmates.com which its purpose it to reunite past school friends.
2. Activity Theory: Activity theory is a philosophical framework that pursues the study of different forms of human practice. From this study it has been realized that the elements of an activity are not static but constantly change based on current social conditions. Thus social conditions and human activities must be analyzed as a whole. It should also be noted that every human develops differently and how people work plays a large part in how they think.
3. Distributed Cognition: Distributed cognition involves the studying of all aspects of cognition. It's the study of more than just the individual and includes interactions socially between people and with materials in the environment. A lot of work within this field has been in the study of such areas as to how to pilot a plane or navigate a ship.
4. Cultural Theories: Cultural theories explore how people relate to their environment and interactions based on their culture. Thus, people of different cultures view various situations much different. For example in some cultures photographs of people are commonly used on web sites to invoke a feeling of trust and comfort where as in other cultures photos of building would work better to serve this purpose. It is important for HCI professionals to not only think of the user when developing an application but to also think of the culture of the user group who might be using this application. It is not uncommon for different versions of the same application to be built depending on the user's culture. If these cultural theories are not taken into account it could result in not only an unusable application but also possibly even an offensive application.
5. Social Informatics: Social Informatics is a study in which the goal is to ensure that the technologies and designs of an application or tool are relevant to the user's life. The focus of this study is to make sure that the design is not technology driven but socially driven. It's important that attention is drawn to functionalities that people find of value and find useful. It's vital to understand who the users are and what there goals are then base the functions of the application off of the user goals.
6. Online Communities: Online communities are built to help communication, bring people together and create a faster, easier method for communicating. Online communities allow people to speak freely about their feelings without the physical barriers that normal communication might place upon them. Online communities can form tight bonds between people. With the rapid creation of Blog sites and various other forums we have seen online communities grow by the thousands. These specific online communities can give individuals a voice of thousands that would normally be quite restricted. When thinking about online communities one might also want to study Social Networks.

Controlled Experiment: In a controlled experiment everything tested remains the same except for the variable (single factor) or variables (multi factor). The goal of this is to make predictions about a behavior and what effects it has on the experiment. When developing a controlled experiment it is best to form a hypothesis (a statement or theory that you believe to be true) and then test this hypothesis. For example, a tabbed style metaphor for navigation is better than plain text links navigation. In this test the navigation type would be the independent variable that would be manipulated to test the hypothesis. The outcomes of the independent variables are the dependent variables because they depend on the treatment of the independent variables.

Single Factor: As the name hints a single factor experiment has only one variable that changes. Because of this it is the simplest experiment to create and test accurately with the smallest amount of analysis. Participants are gathered at random and then randomly assigned to the two groups. If participants are not gathered and selected randomly the tester's bias could interfere with the experiment and skew the results. It is also important to give each user the same instructions as not to lead any particular user. The major draw back to this type of experiment is its efficiency. For example, testing many different variables one at a time could take a lot of time and effort. When a case like this occurs it is better to use a multi factor experiment. This type of experiment might be best used when testing one specific item such as one navigation style against another. The results gained by this test would be very easy to analyze and the test would be very easy, quick and cost effective.

Multi Factor: A multi factor is used to test more than 2 and 3 variables. Anymore than 3 variables can be hard to test accurately and analyze without analytical software. This type of experiment is often used more than the single factor because more research questions can be answered making it more efficient. Within this experiment independent variables can be measured against each other to see what effects are caused. For example, if one wanted to test if men or women could complete an application quicker if the page was either black or white. Both men and women would be tested on black and white pages then analyzed against each other. The gathering of men and women in this example would be a quasi-experimental.

Quasi-Experimental: This means that the groups were not randomly formed. In the example above a selection of men were selected then a selection of women were gathered. It's impossible to assign men and women randomly to a group of men and women, it's already predetermined. Because of this, this group is the easiest to form.

Case Studies: With this type of study one individual is usually focused on in detail. One behavior will be chosen and it will be studied repeatedly. This gives the experimenter a larger amount of control over the situation. However, this can backfire if there was a bias in selecting this individual which can lead to over generalization of other groups or people.