People often leverage artifacts and characteristics from their environment to reduce cognitive load and enhance their cognitive capabilities. External cognition refers to the activities that people use to support their cognitive efforts. These activities rely on: a wide range of artifacts such as computers, watches, pens and papers; characteristics of the environment such as visible landmarks, and signs; and other people. There are three main types of external cognition activities.

These three types of activities are heavily inter-dependent. In the diagram above they are listed from broadest to most specific. The externalization of memory load is the most basic external cognitive activity. It is involved in all types of external cognitive activities.

Computational offloading leverages memory externalization for the specific purpose of performing computational tasks. It is the next most basic external cognitive activity.

Annotation and cognitive tracing can be used to support both types of distributed cognitive activities mentioned above. This type of distributed cognition involves the manipulation or modification of memory and computational externalizations that impact the meaning of the externalizations themselves.

External cognitive activities are used to support experiential and reflective modes of cognition [more info on cognitive modes]. These activities rely on and support all types cognitive processes defined in my earlier post – attention, perception, memory, language, learning, and higher reasoning [more info on cognitive process types].

This framework of external cognitive activities complements the Information Processing model by identifying how people leverage their external environment to enhance and support their cognitive capabilities [more info on information processing model].

It also complements the model of interaction by providing additional insights regarding how people interact with the world (or system images) to support and enhance their cognitive capabilities. However, it does not provide insight into how people interact with systems for non-cognitive pursuits, such as physical and communication ones [more info on model of interaction].

[source: Interaction Design: Beyond Human-Computer Interaction]

** What the hell is ID FMP? **

One of the most prevalent metaphors used in cognitive psychology compares the mind to an information processor. According to this perspective, information enters the mind and is processed through four linear stages that enable users to choose an appropriate response.

Though this model offers insights into how people process information, it is limited by its exclusive focus on activities that happen in the mind. Most of our cognitive activities involve interactions with people, objects, and other aspects of the environment around us. In other words, cognition does not take place only in the mind.

In my next ID FMP post I will cover the external cognition framework that describes external cognitive activities; and distributed cognition models that attempt to map all internal and external activities. Here’s how this model aligns to the frameworks, models, and principles that I have explored over the past several weeks.

The cognitive activities modeled by Information Processing framework above can be mapped to the mental activities outlined in Norman’s Model of Interaction. At a high level, Norman’s model provides additional insights regarding the mental activities that take place and it features the external environment as an important, though unexplored, element. Here is a brief overview of how the phases from this model relates to the interaction one:

• “Input encoding” maps to “perception”
• “comparison” encompasses “interpretation” and “evaluation”
• “response selection” corresponds to “intention” and “action specification”
• “response execution” maps to “execution

The “goals” phase of the Interaction Model crosses over between the “comparison” and “response selection” phases of the Information Processing framework – so do the additional phases of the modified Model of Interaction.

Here is how this model aligns with the framework regarding the relationship between a designer’s conceptual model and a user’s mental model. The focus of the Information Processing model is on the cognitive processes that occur in the user’s mind when they are interacting with the world. These processes are closely related to mental models in two ways:

• First, mental models provide the foundation for people to understand their interactions with the world and select appropriate responses.
• Second, mental models evolve as people evaluate the impact of their own actions and other events on the world.

[Note: by “world” I refer to any physical, virtual and social entities with which people can interact.]

The Conversation Turn Taking Model is related to the Information processing model in a broad sense only. The turn taking framework focuses on explaining an external phenomenon related to language and communication that is driven by the cognitive functions described in the Information Processing model. They do not contradict one another nor do they directly support each other.

The Information Processing model can be applied to both reflective and experiential modes of cognition, though the phases involved in each mode differ. Reflective cognition tends to be active during the “comparison” and “action selection” phases. On the other hand, experiential cognition can be active across all phases depending on the type of interaction.

The chart below provides an overview regarding which cognitive process types are involved with each phase of the Information Processing model.

[source: Interaction Design: Beyond Human-Computer Interaction]

** What the hell is ID FMP? **

There are many theories that attempt to describe the cognitive processes that govern users’ interactions with products systems. Here I will focus on a model developed by Don Norman, which was outlined in his book Design of Everyday Things. This framework breaks down the process of interaction between a human and a product into seven distinct phases.

Seven Phases of Interaction with a Product System

1. Forming the goal
2. Forming the intention
3. Specifying an action
4. Executing the action
5. Perceiving the state of the world
6. Interpreting the state of the world
7. Evaluating the outcome

Two important concepts related to Norman’s Theory of Action are the gulf of execution and evaluation. The gulf of execution refers to the gap between how the user wants to act and how the system allows the user to take action. The gulf of evaluation corresponds to the gap between how the system displays data to how the user interprets this data into knowledge.

Now let’s put this theory into context with some of the concepts and models that we’ve encountered thus far. First, I want to point out that this model aligns with Don Norman’s model regarding the relationship between a designer’s conceptual model and a user’s mental model [read more here]. The focus of this framework is the interaction between the system image, the product’s interface where user interaction happens, and the user’s mental model, the user’s understanding of how the product works which governs the user’s interpretation, evaluation, goals, intention, action specification.

I’ve extended Norman’s original model to account for the reflective cognition that is also involved in peoples’ interactions with products. Reflective cognition governs peoples’ higher-level evaluations, goals and intentions that ultimately drive peoples’ experiential cognition activities. Experiential cognition governs the second-by-second evaluations, goals, and intentions involved in peoples’ interactions with products. These two different modes of cognition are explored in greater detail here.

Here is an example to distinguish and highlight the interdependencies between these two different types of cognition and interaction. Let’s consider a person’s interaction with a car. In this scenario, a person’s reflective cognitive would include setting a goal such as choosing a destination and desired time of arrival, and evaluating what route to take based on understanding of current location and traffic patterns. These activities would govern a person’s experiential interactions with a car and drive their moment-by-moment evaluations, and creation of goals and intentions. Experiential interactions would include using the steering wheel to turn a corner or switch lanes, pressing the accelerator to speed up, and stepping on the breaks to stop the car.

How does the concept of mental models relate to this framework? The mental model itself is not represented by a single phase, or grouping of phases. It refers to the understanding that a user has of how a system works. Norman’s model was developed to describe how users interact with product systems on an experiential, minute-by-minute basis. At this level of interaction a user’s mental model drives their interpretations, evaluations, setting of goals and intentions, and specification of actions.

Now let’s explore how the different cognitive types come into play during the various phases of interaction. These cognitive types have been outlined in greater detail here.

• Attention supports all phases of interaction from perception through to action execution. This cognitive process refers to a user’s ability to focus on both external phenomena and internal thoughts.
• Perception is clearly called out as its own phase in Don Norman’s model.
• Memory plays an important role during all phases from interpretation through to action specification.
• Language supports communication throughout all phases of a person’s interaction with a product. Here I refer to both verbal and visual languages.
• Learning enables people to use new products and increase effectiveness and efficiency in their interactions with existing products. This cognitive process supports all phases between the interpretation and action specification.
• Higher reason governs all activities related to the setting of high-level goals and intent, and driving evaluations.

[source: Interaction Design: Beyond Human-Computer Interaction; and Don Norman's Book The Design of Everyday Things]

** What the hell is ID FMP? **

Holding a conversation is a basic human activity. It requires a large amount of coordination between participants, a fact that is often unnoticed. People need to know when to listen, when they can start talking, and when to cede the floor. Conversation mechanisms facilitate the coordination of conversations by helping people know how and when to start and stop speaking. These mechanisms enable people to effectively negotiate the turn-taking required carry out a conversation.

Harvey Sacks, Emanuel Schegloff, and Gail Jefferson have developed a model that aims to explain how people manage turn taking during conversations. The focus of their research was to create a framework that can be applied across cultures and contexts, and that can accommodate several key observations about the structure and dynamics of conversations. Here is an excerpt from the abstract of their paper The Simplest Systematics for the Organization of Turn-Taking for Conversation.

“The organization of taking turns to talk is fundamental to conversation, as well as to other speech-exchange systems. A model for the turn-taking organization for conversation is proposed, and is examined for its compatibility with a list of grossly observable facets about conversation [outlined below].”

The Foundations
Before we explore the model itself let’s take a look at its foundation. Here is a list of the “grossly observable facets about conversation” that was referred to in the quote above:

1. Speaker changes will always occur and often recur.
2. For most of the time only one party talks at a time.
3. More than one person will often talk at a time, but these occurrences are brief.
4. Most transitions occur with no gap or overlap, or with slight gap or overlap.
5. Turn order varies throughout conversation.
6. Turn size or length usually varies.
7. Length of conversation is not specified.
8. What parties say is not specified.
9. Relative distribution of turns is not specified.
10. Number of parties varies considerably.
11. Talk can be continuous or not.
12. Turn-allocation techniques are used to facilitate the conversation.
13. Sometime turn-constructional units are used to facilitate conversation.
14. Repair mechanisms exist for correcting turn-taking errors.

The Model

The general model that they developed, which is pictured above, is composed of the three basic rules that govern the transition of turns in a conversation. These rules are:

1. The current speaker chooses the next speaker by asking a question or making a request.
2. If the speaker does not choose the next speaker, then another person can self-select to start speaking.
3. The speaker can decide to continue speaking if no other person self-selects to start speaking.

[source: Interaction Design: Beyond Human-Computer Interaction; and Harvey Sacks, Emanuel Schegloff, and Gail Jefferson’s paper The Simplest Systematics for the Organization of Turn-Taking for Conversation, 1974]

** What the hell is ID FMP? **

This assignment was taken from the fourth chapter of the book Interaction Design: Beyond Human-Computer Interactions, written by Helen Sharp, Jenny Preece, and Yvonne Rogers.

Overview
The aim of this activity is for you to analyze the design of a virtual world with respect to how it is designed to support collaboration and communication.

Visit an existing 3D virtual world such as the Palace, habbo hotel, or one hosted by Worlds. Try to work out how they have been designed for taking account of the following:

Assignment Questions
Question A: General social issues

• What is the purpose of the virtual world?
• What kinds of conversation mechanisms are supported?
• What kinds of coordination mechanisms are provided?
• What kinds of social protocols and conventions are used?
• What kinds of awareness information are provided?
• Does the mode of communication and interaction seem natural or awkward?

Question B: Specific interaction design issues

• What form of interaction and communication is supported, e.g. text/audio/video?
• What other visualizations are included? What information do they convey?
• How do users switch between different modes of interaction, e.g. exploring and chatting? Is the switch seamless?
• Are there any social phenomena that occur specific to the context of the virtual world that wouldn’t in face-to-face setting, e.g. flaming?

Question C: Design issues

• What other features might you include in the virtual world to improve communication and collaboration?

Answers
Virtual world selected: Second Life.

Question A
What is the purpose of the virtual world?
According to Linden, Second Life does not have a specific purpose. They describe Second Life as “a free online virtual world imagined and created by its Residents.” Most people use Second Life for entertainment. It enables them to escape to virtual world where then can interact with other real people. It offers an experience that can be likened to the birth child of the SIMS game crossed with a social network. A small segment of Second Life users actually make a living from creating virtual artifacts and owning virtual land.

What kinds of conversation mechanisms are supported?
Second Life supports many of the same conversation mechanisms that people are accustomed to using in real life to govern turn taking. In my personal experience, I continued to follow conversation practices that I am accustomed to using when speaking to someone in person, even though the conversation was taking place on a text-based medium.

The conversation turn-taking model developed by H. Sachs et al. [link] seems to be applicable to this environment (at least according to my very unscientific research). I assume that conversations using voice, which is available in Second Life, support standard conversation mechanisms even more effectively.

Another conversation mechanism that is supported by Second Life is body language. Let me clarify what I mean. Citizens are able select from a large pre-defined list of gestures that enable them to communicate attention, emotion, mood, and more. This is pretty cool feature that can be likened to emoticons on an instant messaging application or social network.

What kinds of coordination mechanisms are provided?
Second Life does a pretty good job here again. They offer robust support for both verbal and non-verbal types of communication. As stated above, users can communicate using a text or voice/audio interface. Avatars are also capable of using a variety of different gestures for communicate. These include nodding yes, or shrugging, clapping, blowing a kiss, and more.

Rules are the foundation of this virtual world on its most basic level. The software code provides a set of rules upon which the entire virtual world is build; these basic rules are documented in the online user guide and help tools. They define the “virtual-physical” world of Second Life, which is the platform upon which user coordination can take place.

One also encounters many rules while exploring the world itself. These external representations are created by users and Linden Lab. They inform other users and help coordinate personal and shared activities. Maps are another key mechanism that supports coordination. They are available to help the users easily locate and transport themselves between islands.

What kinds of social protocols and conventions are used?
Most people seem to mimic real world conventions in Second Life. Conversations are initiated in a manner more akin to real world conversations compared to other types of text-based conversations. Users are conscious of the organization and appearance of the physical artifacts in this virtual world. This is reflected by convention such as the practices of users face one another when speaking, and the fact that many users are extremely conscious of their avatars clothing and style.

What kinds of awareness information are provided?
At the most basic level of awareness, Second Life users are able know who is around them via the visual representation of the virtual world. For the most part, users are able to understand what is happening though this varies considerably based on expertise level. It is possible to overhear others’ conversations as long as they are not having a private chat. Most of the groups of people that I encountered whose physical proximity insinuated that they were having a conversation must have been holding private chats. An interesting design element from the game is how the avatars make a typing movement in the air when they are writing a reply in a conversation.

Does the mode of communication and interaction seem natural or awkward?
The mode of communication and interaction offered in Second Life is natural on most accounts. The natural feel of the text-based conversations is in large part due to our modern-day familiarity holding conversations using messaging applications such as IM and SMS. The overall look and feel of the virtual world is natural. The communicative gestures of the character are fluid and clear in their meaning.

Question B
What form of interaction and communication is supported, e.g. text/audio/video?
Second Life supports all main forms of interaction: text, audio, video, and computational.

What other visualizations are included? What information do they convey?
Second Life is well crafted from a visual perspective. The visual flair is actually provided mostly by the creativity of the members of the community, who develop most experiences and structures that exist in this world. Visualizations that are built into the interface include different modes for displaying chats, maps that provide location information, and the main interface of the virtual world environment.

How do users switch between different modes of interaction, e.g. exploring and chatting? Is the switch seamless?
The switch between different modes of interaction is seamless. If a user is exploring he can easily start chatting with someone else nearby by typing; if a user has a voice-enabled system then they just have to talk. Gestures are not integrated as seamlessly; these have to be selected from a drop-down menu.

Are there any social phenomena that occur specific to the context of the virtual world that wouldn’t in face-to-face setting, e.g. flaming?
As with any medium that allows people to communicate from a distance, people are definitely less concerned with politeness and manners. One social phenomena that I witnessed was a user who kept repeating everything that was said in a conversation between me and a third user.

Question C
Overall, I think that Second Life does a thorough job at providing users with effective communication and collaboration tools. So much so that technology companies such as IBM have built virtual campuses where they hold meetings with employees from around the world. Here are a few ideas that could be explored:

• Allowing users to select moods and emotions. These features would work in a similar way to gestures. The main difference is the duration of a mood or emotion in comparison to a gesture. Moods and emotions last longer and would be controlled using on/off switches.
• Make it easy for users to create and share documents on the fly. Provide capabilities for users to work on documents simultaneously with seamless ability to switch back and forth between focus on the document and on the virtual world.

In my last post I identified two different modes of cognition. Here I will continue my investigation into the scope of cognition by identifying six different types of cognitive processes, taken from the book Interaction Design: Beyond Human-Computer Interaction. My focus will remain on the questions: “what is cognition? And what are the main types cognitive activities?”

The six types of cognitive processes that I will describe are attention, perception, memory, language, learning, and higher reasoning. The processes are interdependent and occur simultaneously. They play a role in experiential and reflective modes of cognition. Here is a description of each process along with a few related implications.

Attention: process for selecting an object on which to concentrate. Object can be a physical or abstract one (such as an idea) that resides out in the world or in the mind.

Design implications: make information visible when it needs attending to; avoid cluttering the interface with too much information.

Perception: process for capturing information from the environment and processing it. Enables people to perceive entities and objects in the world. Involves input from sense organs (such as eyes, ears, nose, mouth, and fingers) and the transformation of this information into perception of entities (such as objects, words, tastes, and ideas).

Design implications: all representations of actions, events and data (whether visual, graphical, audio, physical, or a combination thereof) should be easily distinguishable by users.

Memory: process for storing, finding, and accessing knowledge. Enables people to recall and recognize entities, and to determine appropriate actions. Involves filtering new information to identify what knowledge should be stored. Context and duration of interaction are two important criteria that function as filters.

Design implications: do not overload user’s memory; leverage recognition as opposed to recall when possible; provide a variety of different ways for users to encode information digitally.

Language: processes for understanding and communicating through language via reading, writing, speaking, and listening. Though these language-media have much in common, they differ on numerous dimensions including: permanence, scan-ability, cultural roles, use in practice, and cognitive effort requirements

Design implications: minimize length of speech-based menus; accentuate intonation used in speech-based systems; ensure that font size and type allow for easy reading.

Learning: process for synthesizing new knowledge and know-how. Involves connecting new information and experiences with existing knowledge. Interactivity is an important element in the learning process.

Design implications: leverage constraints to guide new users; encourage exploration by new users; link abstract concepts to concrete representations to facilitate understanding.

Higher reasoning: processes that involve reflective cognition such as problem-solving, planning, reasoning, decision-making. Most are conscious processes that require discussion, with oneself or others, and the use of artifacts such as books, and maps. Extent to which people can engage in higher reasoning is usually correlated to their level of expertise in a specific domain.

Design implications: make it easy for users with higher levels of expertise to access additional information and functionality to carry out tasks more efficiently and effectively.

[source: Interaction Design: Beyond Human-Computer Interaction]

** What the hell is ID FMP? **

Cognition [define] encompasses a wide range of processes related to thinking, sensing, interpreting, evaluating, decision-making, remembering and communicating. It is important for designers to understand human cognition processes in order to design systems that are easy to learn, effective, efficient, pleasurable, and meaningful.

Here, I will first distinguish between two main modes of cognition. In my next post I will identify different categories of cognitive processes. The value of these distinctions is that different modes and types of cognition call for different technology and interaction solutions. It is important to note that both cognitive modes and multiple processes are always active simultaneously.

The focus of this, and my next, post is to explore the scope of cognition. In other words, the question being answered here is “what is cognition? And what are the main cognitive activities?” I will cover models that attempt to illustrate how cognition works at a later time; at which point the question I will address is “how does cognition function?”

The two modes of cognition identified by Don Norman are the experiential and reflective modes. Both of these are essential to human beings, and are continuously used in everyday life often in an overlapping manner. The description below and attached diagram aim to illustrate the main characteristics of each of each modes.

• Experiential: state-of-mind associated to perception of the environment around us, and to our engagement with that environment through our actions and reactions. Contexts where an experiential mode of cognition is used include when a person is having a conversation, driving a car, or reading a book.
• Reflective: state-of-mind associated to higher-level processing of knowledge, memory, and external information (or stimuli) through thinking, comparing, and judging. This type of cognition is needed for people to learn, create ideas, design products, and write books.

[source: Interaction Design: Beyond Human-Computer Interaction; Don Norman's book and Things That Make Us Smart.]

** What the hell is ID FMP? **

We keep on coming back to conceptual models [define]. The reason being, a well-designed conceptual model is a fundamental element of successful product and service systems. To develop a well-articulated conceptual model designers need to think through the main metaphors, concepts, actions, and relationships of the systems they are designing before developing prototypes of any sort (including wireframes, drawings, renderings, etc).

Don Norman’s and Bjoern Hartmann’s model provides insight into how designers’ conceptual models interact and relate to a users’ mental models. It does not, however, provide any guidance to help designer synthesize conceptual models.

Johnson and Herderson’s framework, published in 2002, was developed with this purpose in mind. This framework identifies the standard components of a conceptual model. It provides a blueprint that designer can use to develop conceptual models. Here is an overview of the four components of conceptual model as defined by Johnson and Henderson:

• Major Metaphors and Analogies: Identify important metaphors and analogies used to enable the user to understand what a product does and how to use it.
• Concepts: Define the concepts that users are exposed to and that they need to understand, including the objects the concepts create and manipulate, any relevant attributes, and the operations that can be performed on the concept.
• Relationships and Actions: Identify the relationships between concepts, including whether an object contains another, or is part of it, and the relative importance of objects and actions.
• Mappings: Define the mappings between the metaphors and concepts and the user experience the product is designed to invoke.

Examples of this framework in action (albeit one developed by someone with little to no experience working with it) are available on two of my recent posts:

• The first was developed in response an exercise from the book Interaction Design: Beyond Human-Computer Interaction.
• The second was written as a personal exercise for me to apply this conceptual framework to develop a silly pet product idea that I had been toying around with for a while.

[source: Interaction Design: Beyond Human-Computer Interaction]

** What the hell is ID FMP? **

Developed by Don Norman, the model illustrated below demonstrates how relationship between a designer’s conceptual model and a user’s mental model is mediated by the system image of products or services.

So here is my explanation of what this model means: Designers develop product and service systems based on conceptual models [define] that they create or borrow. I use the terms product and service systems [define] refer to the ecosystem that encompasses products, services and their related artifacts and resources; these can include assets such as manuals and knowledge bases, and resource such as user groups and communities.

Users do not have access to the conceptual models of designers. Their understanding of how a product works is developed based on their interactions with the product itself, their previous experiences with the world, and their existing knowledge and expertise. All of these considerations affect how people interpret their experiences with a product, and the mental model [define] they create to explain how products work.

The term system [define] image refers to the way a product or service system actually appears to a user. System images are always imperfect representations of the conceptual models upon which they were built. For a product to be usable the system image needs to enable users to develop an accurate mental model of how relevant aspects of a product or service works.

An interesting feature of Norman’s 1988 model, is that designers relationship with system images is represented as a one-way phenomenon. This implies that once a product has been designed there is little opportunity for on-going improvements. During the last 20 years advances in technology and design methodology have made it possible for designers to continuously fine-tune product and service systems. This is especially true in the increasingly service-based world of software.

Bjoern Hartmann has revised Norman’s model to reflect the opportunity for designers to play an on-going role in improving the system image of the products they’ve created. The model he proposes includes a feedback loop that enables the user to communicate to the designer via the system image.

Hartmann posits that user-initiated feedback via the system will help identify mismatches between the designer’s conceptual model and the user’s model of how the system functions. Another important consideration is that offering an instantaneous feedback option in the same media on which the interaction is taking place will generate more reliable and richer data than feedback elicited later, or via a different channel.

[Sourced from Don Norman’s website, though I know Norman's framework was also featured in this book The Design of Everyday Things; Paper by Bjoern Hartmann written during graduate studies at Standford]

** What the hell is ID FMP? **

As I embark on my exploration of frameworks, models, and principles related to interaction and experience design I will maintain this working list of definitions regarding key terms and concepts.

This is by no means a comprehensive list. This is actually an utterly selfish endeavor as these definitions are solely intended to help me in my own study of this field. They have been cherry-picked, mostly from online sources and the book Universal Principles of Design.