Human Computer Interaction Paper 2022 October

Question 1

i) Identifying User Types

There are two main user groups of a digital token-based meal allocation system:

• Students: Students who are deemed 'needy' are the recipients of the meal tokens and can redeem them at the canteen.
• Canteen Staff: Canteen staff are responsible for issuing the allocated amount of food upon presentation of valid tokens and managing the token system.

ii) Building a Student Persona

Name: Sarah

Age: 19

Background: Sarah is a first-year university student from a low-income family.

Goals: Sarah's primary goal is to receive nutritious meals without experiencing financial burden. She also wants a system that is discreet and doesn't single her out as 'needy'.

Needs: Sarah needs a system that is easy to use and understand. It's important that she can access her meal tokens and check her balance easily. A system that offers some flexibility, like occasional token sharing with friends facing temporary financial constraints, would be beneficial.

Frustrations: Sarah wants to avoid situations where she's in line and realizes her balance is insufficient or experiences difficulties with the token redemption process, especially during peak hours.

Explanation: This persona represents a typical user by focusing on the financial challenges and desire for discretion. The emphasis on system usability and potential token sharing highlights the need for a user-friendly and somewhat flexible system.

iii) Global Structure of User Interface

A three-level global structure for the customer (student) interface could be:

Level 1: Home

• View Meal Balance
• Redeem Tokens
• Transaction History

Level 2: View Meal Balance

• Current Token Balance
• Recent Transactions

Level 2: Redeem Tokens

• Enter Token Amount
• Confirm Redemption

Level 2: Transaction History

• Filter by Date
• View Details

iv) Considerations for Token Sharing

Introducing token-sharing functionality requires careful consideration of potential challenges:

• Fairness: Establish clear guidelines and limits on sharing to prevent exploitation and ensure equitable meal access for all needy students. For example, implement a cap on the number of tokens transferable per period or limit transfers to verified 'needy' students only.
• Security: Implement robust authentication measures to prevent unauthorized sharing or token theft. Consider requiring two-factor authentication or biometric verification for sharing transactions.
• Transparency: Maintain a transparent transaction history visible to both the sender and receiver. This accountability can deter misuse and build trust in the system.

v) Interface and Interaction Strategy

• Streamlined Redemption Process: Implement a system that minimizes steps for token redemption, possibly through QR code scanning or NFC technology. The system should display clear confirmation messages after each transaction with details about the remaining balance.
• Offline Functionality: Offer offline token access to ensure meal availability even when network connectivity is unreliable. Explore options like temporary QR codes or offline balance storage.
• Clear Error Handling: Design the system to display clear and informative error messages in case of invalid tokens, insufficient balance, or system errors. Provide guidance on resolving issues to prevent user frustration.
• User Feedback Mechanism: Integrate a feedback system for users to report issues, suggest improvements, or seek clarification. Regular review of feedback can enhance system usability and address user concerns.

By implementing these strategies, the digital token-based meal allocation system can effectively address the needs of all stakeholders while ensuring an efficient and positive user experience.

Question 2

i) Evaluating the Interface Design of the University of Moratuwa Research Office Website

Positive Aspects:

• Clear Branding: The website features the University of Moratuwa logo prominently at the top, establishing a clear visual identity.

Negative Aspects:

• Lack of Visual Hierarchy: The website lacks a clear visual hierarchy, making it difficult for users to scan and find the information they need. The text is presented in a dense and overwhelming manner. This goes against Shneiderman and Plaisant’s golden rule of striving for consistency and Nielsen's heuristic of 'aesthetic and minimalist design'.
• Text-Heavy Content: The website relies heavily on text, which can be overwhelming for users. Breaking up the text with visuals, headings, and bullet points would improve readability.
• Poor Navigation: The navigation menu is cluttered with too many links that are not well-organized, making it difficult for users to find specific information.

ii) Mobile Interface Design

A mobile-first approach is crucial for modern web design. Here’s a suggested mobile interface design for the webpage:

• Hamburger Menu: Implement a hamburger menu to consolidate the navigation options.
• Visual Hierarchy: Prioritize important information and use clear headings and subheadings to break up text.
• Concise Content: Optimize text for mobile consumption, using bullet points and shorter paragraphs.
• Touch-Friendly Design: Ensure buttons and links are appropriately sized for touch interaction.

Justification: This design enhances the user experience by:

• Improved Navigation: The hamburger menu declutters the screen and makes navigation intuitive on smaller devices.
• Enhanced Readability: Clear visual hierarchy and concise content make it easier to read and find information.
• Ease of Use: Touch-friendly design ensures effortless interaction on touchscreens.

iii) Redesigning the Interface for Improved UX

Design Sketches:

Justifications:

• Simplified Navigation: The redesigned interface should have a clear and concise navigation menu with logical categories and subcategories. A mega menu could be employed for better organization of the "More Links" section.
• Visual Hierarchy: Use a clear visual hierarchy with headings, subheadings, bullet points, and images to make the content more scannable and digestible.
• Whitespace and Typography: Incorporate ample whitespace to reduce visual clutter and improve readability. Choose a legible font and appropriate font sizes for different screen sizes.
• Responsive Design: The website should be responsive, automatically adjusting its layout and content to different screen sizes, ensuring a seamless experience across desktops, laptops, tablets, and smartphones.

iv) User Customizable Interface and Interaction

Recommendation: While a fully customizable interface might be overly complex for this specific website, offering some level of personalization could enhance the user experience.

Reasons:

• Personalized Experience: Allowing users to prioritize content or save preferences can create a more tailored experience.

Technical Realization:

• User Profiles: Implement a simple user profile system where users can choose their interests or frequently accessed sections.
• Cookies: Use cookies to remember user preferences, such as language settings or theme preferences.
• Content Filtering: Allow users to filter content based on keywords or categories.

Question 3

Question 3 i)

Universal usability is the ability of a system to be used by anyone, regardless of their physical or cognitive abilities.

Example: Smart home systems

Human Aspect: A diverse user base interacts with the smart home system. This includes individuals with varying ages, technical expertise, physical abilities, and cognitive capabilities. For instance, a user might be an elderly person with limited mobility, a child learning to interact with technology, or an individual with visual impairments.

Computer Aspect: The smart home system comprises various interconnected devices. These devices may include sensors (for motion, temperature, light), actuators (smart locks, smart lights), and a central hub for control and automation.

Interaction Aspect: Users interact with the smart home system through multiple modalities. These modalities encompass:

• Voice commands: Users can issue verbal instructions to control devices. For example, a user can say, "Turn on the living room lights".
• Mobile applications: User-friendly apps on smartphones or tablets provide a visual interface for controlling and monitoring the system. This allows users to adjust settings, schedule actions, and receive notifications.
• Gesture recognition: Some systems allow users to interact with gestures, like waving a hand to turn on lights. This is particularly beneficial for individuals with limited mobility.
• Sensors and automation: The system can automatically respond to environmental changes. For instance, lights can turn on at sunset or the thermostat can adjust based on room occupancy.

Question 3 ii)

To further enhance universal usability by incorporating multi-modality into the smart home system, one could consider:

• Integrating with wearable devices: Allow users to control aspects of the smart home system using smartwatches or fitness trackers. This could include receiving notifications or controlling devices directly from the wearable.
• Brain-computer interfaces (BCIs): Although still in its early stages, BCI technology holds the potential for direct thought-based control of smart home systems, particularly beneficial for individuals with severe disabilities.
• Haptic feedback: Incorporate haptic feedback devices into the system. For example, a user could receive a subtle vibration on their wristwatch to confirm a command or be alerted to a notification.

Question 3 iii)

To facilitate universal usability within my Final Year Project software product, which aims to create a framework for large language models to enable normative reasoning, I could consider incorporating various interfaces and interaction options inspired by the diverse applications of LLMs in HCI. Here are some design ideas and their justifications, keeping in mind the goal of universal usability:

• Natural Language-Based Interaction:
  • Interface: A chat-like interface would allow users to interact with the system using natural language, similar to how users interact with chatbots.
  • Wireframe Example: A simple chat window where users can type in their queries or requests in natural language and receive responses in a similar format.
  • Justification: LLMs excel at understanding and generating human-like text, making such an interface intuitive and accessible for users with varying technical expertise. This aligns with the principle of "Match between system and the real world," as highlighted by Nielsen's heuristics, by using familiar language and interaction patterns.
• Voice-Based Interaction:
  • Interface: Integrate voice input and output capabilities, allowing users to interact with the framework using voice commands and receive audio responses.
  • Wireframe Example: A representation of a microphone button that users can press to initiate voice commands, and a speaker icon to indicate audio output.
  • Justification: This caters to users who prefer spoken language or have accessibility needs related to typing or reading. Enhanced voice assistants powered by LLMs can significantly improve user experience by offering more natural and efficient interactions. This aligns with Shneiderman and Plaisant's golden rule of "Cater to universal usability" by providing alternative input methods.
• Gesture Recognition:
  • Interface: If the framework involves a visual environment (e.g., for data visualization or interaction with normative scenarios), explore integrating gesture recognition as an input method.
  • Wireframe Example: A representation of a virtual hand performing gestures, such as pointing, selecting, or manipulating objects within the framework's visual environment.
  • Justification: Gesture-based controls offer a more natural and intuitive way to interact, especially within spatial computing environments. This approach can be further enhanced by integrating Shneiderman and Plaisant's principle of "Strive for Consistency" in standardizing gesture controls across different platforms.
• Adaptive User Interfaces:
  • Interface: The framework could dynamically adjust the complexity of the interface or the level of information displayed based on the user's proficiency and context.
  • Wireframe Example: Two versions of the same screen - a simplified version with fewer options and a more detailed version with advanced settings. The system could switch between these based on user interaction.
  • Justification: Tailoring the interface to the user's needs can make the system more approachable for beginners while still providing powerful features for experts. This aligns with Nielsen's heuristic of "Flexibility and efficiency of use" by catering to a wider range of user expertise and preferences.
• Multimodal Interaction:
  • Interface: Combine multiple input modalities, such as natural language, voice, and gestures, to provide users with flexibility and choice in how they interact with the framework.
  • Wireframe Example: A screen with different sections for voice input (microphone icon), text input (chatbox), and a visual area where users can interact using gestures (represented by a virtual hand).
  • Justification: Multimodal interfaces cater to diverse user preferences and abilities, making the system more inclusive. Integrating feedback mechanisms, as highlighted in Shneiderman and Plaisant's golden rules, is crucial to ensure effective multimodal interaction.

Justifications based on Universal Usability:

• Accessibility: Provide options for users with disabilities. For example, text-to-speech and speech-to-text functionalities can be crucial for users with visual or motor impairments. The use of clear and concise language, adjustable font sizes, and sufficient colour contrast are also vital considerations.
• Learnability: The framework should be easy to learn for new users, regardless of their prior experience with LLMs or normative reasoning. This can be achieved through intuitive design, clear instructions, and guided tutorials.
• Efficiency: Enable experienced users to interact with the framework quickly and efficiently. Features like keyboard shortcuts, command history, and advanced filtering options can enhance user productivity.
• Error Prevention and Recovery: Minimize the potential for user errors and provide clear guidance on how to recover from them. This can involve incorporating confirmation dialogues for critical actions, clear error messages with actionable suggestions, and undo/redo functionalities.

By carefully considering these interface and interaction design aspects, you can contribute to a more usable and inclusive framework that accommodates a wider range of users, ultimately promoting the understanding and application of normative reasoning within LLMs.

Question 4

• i) Mixed Reality

  Mixed Reality (MR) combines elements of Augmented Reality (AR) and Virtual Reality (VR) to merge the real and virtual worlds, allowing interaction between physical and digital objects. This means that digital content is not merely overlaid onto the real world (as in AR) nor does it completely replace it (as in VR). Instead, MR allows digital content to interact with and be anchored to the real world in a way that feels seamless and natural.

  One example is Magic Leap, a device and platform that blends the real and virtual worlds. Users can see and interact with digital objects, like dragons flying in their living room, that appear to exist in their physical space.

• ii) Ubiquity and Semantic Web

  • Ubiquity: In the context of Web 3.0, ubiquity refers to the omnipresence of the internet and digital information. Web 3.0 envisions a world where access to information and functionality is not limited to specific devices or locations. Instead, users can seamlessly transition between devices and environments while staying connected. An example application of this concept is the smart home. Users can control appliances, lighting, and entertainment systems through interconnected devices and voice commands, regardless of their location within the home.

  • Semantic Web: The semantic web focuses on making internet data understandable to machines. Web 3.0 aims to move beyond keyword-based searches to a web where machines can understand the meaning and context of data. This enables more intelligent search results, personalised recommendations, and sophisticated data analysis. An example of this is Quickpod, an application that aims to enhance news consumption. Quickpod utilises AI to analyse user data, interests, and demographics to offer personalised news articles and content recommendations, streamlining the information intake process.

• iii) Brain-to-Computer Interfaces Application Scenarios

  • Restoring Communication: BCIs offer a lifeline for individuals who have lost their ability to speak or move due to conditions like ALS or brainstem stroke. By decoding brain signals related to intended speech or movement, BCIs can enable these individuals to control communication devices, such as virtual keyboards or speech synthesisers, using only their thoughts. This restores their ability to express themselves and interact with the world.

  • Controlling Prosthetics: BCIs have the potential to revolutionise prosthetic limbs for amputees. By implanting electrodes in the brain's motor cortex, BCIs can interpret neural signals associated with intended movements, allowing users to directly control the movements of their prosthetic limbs. This level of control offers significantly improved dexterity and functionality compared to traditional prosthetics, restoring a sense of agency and independence for users.

• iv) Kinaesthetic Communication

  Kinaesthetic communication refers to communication through body language and physical movement. It encompasses gestures, facial expressions, posture, and overall body movement to convey messages.

  An application that relies heavily on kinaesthetic communication is virtual reality (VR). In VR environments, users interact with the virtual world using their body movements, which are tracked by sensors. For example, a user might reach out with their hand to grab a virtual object or turn their head to look around the environment.

• v) Virtual Reality Exposure Therapy (VRET) Usages:

  • Treating Phobias: VRET effectively addresses phobias like fear of spiders, heights, or public speaking by creating realistic virtual environments where individuals confront their fears in a controlled and safe setting. Repeated exposure to the feared stimulus helps desensitise the individual and reduce their anxiety levels.

  • Post-Traumatic Stress Disorder (PTSD): VRET can be beneficial in treating PTSD by creating simulations of traumatic events, allowing individuals to process the trauma in a controlled environment. This exposure can help individuals manage their emotional responses and cope with traumatic memories.

  • Anxiety Disorders: VRET helps individuals with anxiety disorders like social anxiety or generalized anxiety disorder by exposing them to anxiety-provoking situations in a virtual setting. This allows them to practice coping mechanisms and gradually reduce their anxiety in real-world situations.

• vi) Immersive 3-D Experience Products/Technologies

  • Oculus Quest 2: A standalone VR headset that provides immersive experiences without requiring a PC or console. It offers high-resolution displays, intuitive controls, and a wide range of applications and games for an immersive 3D experience.

  • Apple Vision Pro: This augmented reality (AR) headset combines AR and VR to create a mixed reality experience. Its high-resolution displays, advanced sensors, and eye and hand-tracking capabilities enable users to interact with 3D digital content in their physical environment.

  • Meta Quest Pro: Building upon the success of the Oculus Quest, this more advanced version features higher resolution, improved ergonomics, and enhanced tracking capabilities. It caters to both gaming and professional use cases, delivering immersive 3D experiences.