Foundations | Module 3 - Human Visual Perception (Mod. 3) Flashcards
Created using my notes and by adding the lecture videos in Notebook LM and asking for a study guide.
Which technique is most effective for mapping categorical data to colors in data visualizations, ensuring clarity, consistency, and easy interpretability?
1. Applying gradient color scales
2. Randomly assigning colors to categories
3. Utilizing a pre-defined, structured color palette
4. Limiting color usage to only primary colors
3) Utilizing a pre-defined, structured color palette
A pre-defined, structured color palette ensures clarity, consistency, and accessibility. It helps viewers quickly distinguish categories and maintains uniformity across visualizations, especially when dealing with multiple datasets.
What is visual perception?
In data visualisation, it refers to the process by which individuals interpret and understand visual information presented to them. It plays a crucial role in how we perceive data and make sense of visual representations. Here are some key aspects:
Interpretation: Visual perception allows us to interpret patterns, trends, and relationships within data through graphical representations.
Human Visual System: Understanding how our eyes and brain work together to process visual stimuli helps in designing effective visualizations that are easy to comprehend.
Encoding Data: Different visualization techniques (like charts and graphs) use visual variables (such as color, size, and shape) to encode data, making it easier for viewers to grasp complex information quickly.
Cognitive Load: Effective visualizations minimize cognitive load, allowing viewers to focus on the data rather than struggling to understand the visual format.
Why is understanding human visual perception important in the context of data visualisation?
Understanding human visual perception is crucial because data visualisations present graphics that are interpreted by our eyes and brain. By considering how we visually process information, designers can create more effective and accurate representations of data, ensuring the intended insights are clearly conveyed. For example:
Effective Use of Color: highlight important data points, differentiate categories, and guide the viewer’s attention to key insights.
Appropriate Chart Types: Different data types require different visualization techniques. Knowing how viewers interpret various chart types allows designers to choose the most effective representation.
Visual Hierarchy: Establishing a visual hierarchy through size, contrast, and layout helps direct the viewer’s focus to the most important elements of the visualization, making it easier to understand the main message.
Grouping and Patterns: Recognizing how humans group visual elements can help in designing visuals that effectively communicate relationships and patterns within the data, making it easier for viewers to draw conclusions.
Why is it essential to make strategic choices when selecting and applying visual variables in data visualizations?
- To simplify the design process and speed up visualization creation
- To minimize time and effort during the design and development stages
- To clearly communicate key data insights and improve the interpretability of the visualization
- To prioritize aesthetic appeal and increase visual sophistication
To clearly communicate key data insights and improve the interpretability of the visualization, making it easier for viewers to understand and interpret the visualization.
Which of the following is considered one of the foundational “laws of grouping” that guide the perception of data in visualizations, according to Gestalt principles?
1. Law of Closure
2. Law of Continuity
3. Law of Proximity
4. Law of Similarity
Law of Similarity
The Law of Similarity states that elements that share similar visual characteristics (such as color, shape, or size) are perceived as part of the same group or category, helping to establish relationships between data points and facilitate pattern recognition.
Which of the following visual encoding techniques is most suitable for representing categorical data, ensuring clear differentiation between distinct groups?
- Color hue
- Position along a common axis
- Size of shapes (Area or length)
- Shape orientation (Angle or rotation of shapes)
1) Color hue
It’s one of the most effective encoding techniques for distinguishing categorical data. By assigning distinct hues (e.g., red, blue, green) to different categories, you can easily differentiate between them and highlight their relationships, especially when there are multiple categories to represent.
What is the primary advantage of using size as a visual encoding technique in data visualization?
It provides an intuitive and effective representation of quantitative differences
Size is a powerful visual encoding technique that allows for a direct and proportional representation of quantitative differences. Larger sizes correspond to larger values, making it easy for viewers to compare data points, spot trends, and identify outliers in the dataset.
Which of the following chart types is most effective for visualizing the correlation between two continuous variables in a dataset?
1. Scatter plot
2. Pie chart
3. Histogram
4. Bar chart
1) Scatter plot
This is the best choice for visualizing the correlation between two continuous variables, as it allows you to observe patterns, trends, and potential outliers in the data.
What is the main purpose of using color in data visualizations?
To help distinguish between data categories and highlight important trends or patterns
Color makes it easier to differentiate data, enabling users to easily identify categories, patterns, and key insights for better analysis and decision-making.
Why is it essential to have a deep understanding of human visual perception when designing effective data visualizations?
It aids in selecting colors and visual cues that are easily distinguishable by the human eye.
Understanding human visual perception enables designers to choose colors, shapes, and visual elements that are easy to differentiate. This supports intuitive interpretation, reduces cognitive load, and helps users quickly identify key insights in the data.
Discuss a potential challenge associated with using “size” as a visual variable to encode quantitative data. Provide an example to illustrate your point.
A potential challenge with using “size” is that our perception of size can be easily influenced by surrounding elements or patterns, leading to inaccurate comparisons. For example, two circles of the same size might appear different if one is surrounded by smaller circles and the other by larger ones.
Why should the “rainbow” colour map generally be avoided in data visualisation?
The “rainbow” colour map is generally avoided because it is perceptually uneven, meaning that changes in colour do not consistently correspond to changes in the underlying data values. This can introduce misleading information and make it difficult for viewers to accurately interpret the data
Describe briefly the main 8 visual variables with an example for each.
1) Position: This refers to the location of a visual object. For instance, as age increases, a mark might appear further along a 1D axis. On a map, different phenomena can be represented by marks at different geographical locations.
2) Marks: These are the basic graphical elements or primitives used in a visualisation, often called glyphs or symbols.
3) Size: This visual variable can represent data through changes in the dimensions of a mark. This can be a 1D change (length), 2D (area) or 3D (volume). An example is associating a quantitative variable to the size of bubbles in a bubble chart.
4) Brightness: This refers to the luminance or how light or dark a visual element appears. An example provided is the difference in brightness between two boxes, although the source notes that perceiving precise differences in brightness can be challenging compared to size.
5) Colour: This visual attribute involves different hues, saturations, and luminances. Colour maps are used to map variables to colours. For continuous variables like ratios, gradients of colour can be used. For discrete or categorical data, palettes of distinct colours are more appropriate. For ordinal data, palettes that follow a logical order should be used.
6) Orientation: This involves mapping data to the angle or direction of a shape or mark. An example is visualising wind speed where the orientation of an L-shaped mark indicates the direction of the wind. The source highlights the importance of considering symmetry when using orientation, as a simple line might be ambiguous in terms of direction.
7) Texture: This refers to the surface pattern of a visual element. It is noted that texture is more commonly used in black and white graphics. Texture can be used to differentiate data or clusters, but care must be taken when encoding ordinal data with texture as it can be difficult to perceive a clear order.
8) Motion: This visual variable uses animation to represent changes in data over time or to show interpolation between different values. An example is showing how different views in a visualisation update in a synchronized manner using motion.
Briefly describe the visualisation pipeline, highlighting the key stages involved in creating a visual representation of data.
The visualisation pipeline involves several stages:
1. raw data
2. data analysis
3. data preparation (filtering/focusing)
4. mapping data to geometric data (VISUAL ENCONDING)
5. rendering the information to create an image.
The mapping stage, or visual encoding, is particularly important as it determines how data attributes are converted into visual elements.
Briefly define the Gestalt principle of “proximity” and explain its implication for designing effective visualisations
The principle of “proximity” states that elements that are positioned close together are perceived as belonging to the same group. In visualisation design, this implies that related data points should be placed closer to each other to visually indicate their association and avoid unintended groupings due to spacing.
How does the Gestalt principle of “similarity” influence how viewers perceive groups of visual elements in a data representation?
The principle of “similarity” suggests that elements that share similar visual attributes (e.g., shape, colour) tend to be perceived as a unified group. This principle can be used in visualisations to group data points belonging to the same category or sharing common characteristics, aiding in pattern recognition.
Describe what is visual enconding.
Visual encoding is the core process of mapping data attributes and their relationships to a visual representation.
In other words, it’s strategic decision-making process of choosing which visual variables (like size, colour, position, etc.) will represent which data attributes to effectively communicate the underlying information.
Describe what is a mapping function. Provide a couple of examples of such functions.
A mapping function in the context of data visualisation is an algorithm or set of rules that defines how data attributes are converted into a visual representation. This algorithm can be implicitly or explicitly defined by the designer.
For example, a mapping function might specify that the ‘sales’ data attribute will be represented by the length of a bar in a bar chart (using the visual variable of size/length), or that different ‘product categories’ will be represented by different colours (using the visual variable of colour).
In data visualisation, when does it make sense to think in terms of mapping functions?
The concept of a mapping function is more relevant for repeatable visualisations. In such cases, the mapping needs to consider requirements, edge cases, outliers, and scalability.
This suggests that the mapping function should be robust and consistently translate data to visuals under various conditions.
What is the relationship between visual enconding, visual variables and the Gestalt laws of grouping?
The Gestalt laws are critical to consider during the visual encoding process when choosing and manipulating visual variables because they significantly influence how viewers will perceive the relationships and structures within the data.
What are the Gestalt laws of grouping? Provide examples of them.
These laws are a set of principles in psychology that explain how humans naturally perceive objects as organised sets of patterns and objects. These laws describe how our visual system automatically groups together elements that share certain visual characteristics.
1) Proximity: Elements that are close together in space tend to be perceived as belonging to the same group. For example, a set of dots spaced closely horizontally will be seen as rows, while the same dots spaced slightly closer vertically will be seen as columns. If dots are clustered into two distinct spatial areas, we will perceive two separate groups. This implies that in visualisation design, related information should be placed closer together, and the spacing between elements can influence the perceived groupings.
2) Similarity: Elements that are similar to each other in appearance (e.g., shape, size, colour) tend to be perceived as a unified group. The sources illustrate this by showing that similar objects are grouped together regardless of slight differences in colour or the symbols used. This suggests that using similar graphics can effectively define rows, columns, and other groupings of marks in a visualisation.
3) Connectedness: Elements that are visually connected to each other are perceived as being more related than elements that are not connected, even if they are close together or similar. The sources state that connectedness is typically more powerful than proximity and similarity in forming groups. Connectors can be used in visualisations to link groups and marks, but care should be taken to avoid making the visualisation too cluttered. The sources mention that our minds naturally prefer to interpolate smooth shapes for connections.
4) Symmetry: Elements that are symmetrical to each other tend to be perceived as a unified group and are more likely to be clustered together than asymmetrical objects. An example given is an image with both black and symmetrical blue shapes; after looking at it for a short time, viewers tend to focus more on the blue shapes due to their symmetry. This suggests that using axes and other frames of reference that create symmetry can support the intended interpretation of a design.
5) Closure: We tend to perceive closed contours or complete figures even when they are not entirely present. For instance, we will see a circle even if there are small gaps in its outline. The implication for visualisation design is that occluding shapes can lead to incorrect assumptions as viewers might mentally close incomplete forms.
6) Figure and Ground: We tend to organise visual information into a foreground (figure) and a background (ground). Smaller parts of a pattern are typically perceived as the foreground. The sources illustrate this with examples where smaller patterns within a larger boundary are seen as the object of focus. This implies that smaller areas within larger boundaries in a visualisation will likely be the first elements users attempt to interpret.
