IV. Visual encoding - marks and channels

1. Creating visualization idioms

Encoding relevant data and map onto visual encodings (=Marks and Channels)

Marks

Marks are geomatric primitive objects classify according to the number of spatial dimensions they require.

For example:

• a zero-dimensional (0D) mark is a points
• a one-dimensional (1D) mark is a line
• a two-dimensional (2D) mark is an area

Channels

• Parameters control the appearance of marks
• The same information can be mapped onto mutiple channels

→ All channels are not equal: the same data attribute encoded with two different visual channels will result in different information content.

• Channel Types
  • The identity channels tell us information about what something is or where is it
  • The magnitude channgels tell us how much of something there is

2. Principles: Expressiveness and Effectiveness

The use of marks and channels in vis idiom design should be guided by the principles of expressiveness and effectiveness. These ideas can be combined to create a ranking of channels according to the type of data that is being visually encoded.

2.1. Expressiveness principle (match channel and data characteristics)

The visual encoding should express all of and only the information in the dataset attributes Thể hiện tất cả và chỉ thông tin có trong thuộc tính dữ liệu

• Ordered data should be shown in a way that our perceptual system intrinsically senses as ordered Dữ liệu có thứ tự nên được thể hiện sao cho chúng ta tự nhiên nhận thức được là có thứ tự
• Unordered data do not require this.

→ The identity chanels are the correct match for the categorical attributes. The magnitude channels are the correct match for the ordered attribute, both ordinal and quantitative

2.2. Effectiveness principle (encode most important attributes with highest ranked channels)

• The importance of the attribute should match the salience (= noticeable / nổi bật) of the channel.
• Effectiveness of different channels depends on human factors

→ The most important attributes should be encoded with the most effective channels in order to be most noticeable, and then descreasingly important attributes can be matched with less effective channel

→ What the word effectiveness means in the context of visual encoding? According to the criterial of accuracy, discriminability, separability, popout and grouping

2.2.1. Accuracy

• Perceptual system mostly operates with relative judgements, not absolute.
• That’s why accuracy increases with common frame/scale and alignment
• Weber’s Law: ratio of increment to background is constant
  • filled rectangles differ in length by 1:9, difficult judgement
  • white rectangles differ in length by 1:2, easy judgement

2.2.2. Discriminability

The principle of discriminability refers to the ability of a visual channel to effectively differentiate between items in the data

(Discriminability đề cập đến khả năng phân biệt các sự khác biệt giữa các đối tượng hoặc giá trị khi sử dụng một kênh trực quan)

• Ability to perceive differences in mapped variables

  

• Depends on

  • Channel features
  • But also on layout, color, size, cardinality, etc.
  • Human can distinguish small differences in local comparisions but not at a distance

2.2.3. Separability

The separability principle in visual encoding refers to the independence or interaction between visual channels used to represent different attributes of data.

(Separability đề cập đến khả năng biểu thị các thuộc tính khác nhau mà không có sự tương tác hoặc can thiệp giữa chúng.)

• Separable: if they can be perceived as two distinct values

  → Use separable channels to perceive one variable at a time

• Intergal: If the feature can be easily perceived only as a one

  → Use integral channels to obtain a holistic effect

2.2.4. Popout / Saliency

Do you see the black sheep?

Where is Wally?

• Saliency is the degree to which a specific location or region in an image stands out and attracts attention.

• Attention has a fundamental role in visualization. Humans do not perceive all the visual information in front of them

  • Eyes are constantly moving and we focus only a small region of the images at each time
  • Part if the visualized information may be not perceived by the users
  • There are solutions that are more effective to avoid the risk

• It seems that we have two different visual mechanisms

  • Pre-attentive

    • Less than 200 ms: before human vision starts to search.
    • Parallel on the visual field, independent on the number of distractors.
    • Speed depends on channel and amount of difference from distractors

    → True for many individual channel/graphics primitives

  • Attentive: more time required, requires sequential search, time grows with the number of distractors

    → True for most combinations of channels

<aside> 📒

Notes

• Preattentive stimuli may have different strenghts (color has faster pop up than size)
• May be asymmetric (big particles in a set of small pop out faster than small particles in a background of big ones) </aside>