Using science to guide design - "Cognition and Design" reading notes

[Format]

Original excerpt

In addition to vision, our goal is to filter perception from other senses. A familiar example is the "cocktail party" effect. If you are talking to someone at a crowded cocktail party, you can focus most of your attention on what they are saying, even if there are many other people talking around you. The more interested you are in the conversation, the more you filter out the conversation around you. If you get bored of the conversation, you'll probably start hearing more and more of the conversation around you.

Target filtering of perception is particularly reliable in adults, who are more focused on targets than children. Children are more easily driven by stimuli and have fewer goals filtering their perceptions. This characteristic makes them more easily distracted than adults, but also makes them less susceptible to being influenced when observing.

In contrast, writing and reading did not appear until several thousand years ago before the B.C.E., and did not become common until four to five hundred years ago, well after the human brain reached the modern level of evolution. As children, our brains do not display any special innate ability to read. Reading is actually a man-made ability acquired through systematic instruction and training, just like playing the violin, juggling, or reading music (Sousa, 2005).

In other words, the most effective reading method is a context-free, bottom-up feature-driven method, which requires mastery to the unconscious level. Although it is two parallel reading methods with feature reading, context-driven reading is now mainly regarded as a candidate method, which only works when feature-driven reading has difficulties or cannot achieve sufficient unconsciousness.

Why is context-free (bottom-up) reading not unconscious in some adults? This continues to intensify when some people do not gain enough reading experience in childhood, allowing feature-driven identification processes to become unconscious. They find reading mentally strenuous and stressful as adults and therefore avoid reading. to address their deficiencies in reading ability.

For example: "Your session has expired. Please re-authenticate.". Most non-technical users do not understand the meaning of "re-authentication" and exit the unconscious reading state.

Omitted.

If you have taken psychology or neurophysiology courses in school, you may already know that the retina in the eye (that is, the surface on which images are focused on in the eyeball) has two types of photoreceptor cells: optic cells and Optic cone cells. You may also know that rod cells detect light intensity but not color, while cone cells detect color. Finally, you may also know that there are three types of cone cells that are sensitive to red, green, and blue light. This means that our color vision is similar to that of cameras and computer monitors, with the combination of red, green, and blue pixels detecting various forms. color.

First of all, we in industrialized societies hardly use rod cells, which only work at low brightness. They only work in very low-light environments, such as those our ancestors lived before the 19th century. Today, we only use our rods during candlelight dinners, fumbling around dark rooms at night, and camping out at night. In bright daylight and artificial lighting (where we spend most of our time), the rod cells are completely overexposed and provide no useful information. Most of the time, our vision is based entirely on information provided by the cells in the optic cone.

(Introduced separately below) Optic cone cells:

As shown below:

Our vision is optimized for detecting contrast (edges), not Absolute brightness.

Being sensitive to contrast rather than absolute brightness is an advantage of humans, so that our primitive ancestors can separate and hide nearby whether it is a sunny noon or a cloudy morning. The bushes contain leopards and other similar dangerous animals.

Likewise, being sensitive to color contrast rather than absolute color causes us to perceive a rose as equally red in the sun as in the shadow.

The fovea occupies only 1% of the retinal area (hold your arm straight and touch your thumb. Your thumbnail, viewed from one arm away, is about the same diameter as the fovea) (comparable in size), while 50% of the visual cortex of the brain is used to receive input from the fovea. Progress says that the connection ratio between the optic cone cells in the fovea and the ganglion cells, the starting point of visual information processing and conduction, is 1:1, while in the rest of the retina, multiple photoreceptor cells (optical cone cells and rod cells) Connected to a ganglion cell.

The resolution of normal people's fovea is very high: they can distinguish thousands of points in that area, which is higher than many pocket digital cameras today. And our visual resolution of borders is so poor.

Then someone must ask why the world we see is not a tunnel, that is, everything except what we are looking at directly is out of focus. On the contrary, everything around us is also clear. . We feel this way because our eyes are constantly moving at a rate of about three times per second, selectively focusing on objects in the surrounding environment in a rough, impressionistic way, based on what we know. and, as expected, fill in the rest of the field of view. The brain doesn't need to maintain a high-resolution mental model of our surroundings because it can instruct the eye to sample and resample specific details as needed.

Peripheral vision exists primarily to provide low-resolution cues to guide eye movements, allowing the center to see all that is interesting and important in the field of view, or to detect movement in another part of our field of vision. An interesting fact is that it has a gap, a small area where we see nothing. This gap corresponds to the exit of the optic nerves and blood vessels in the retina behind the eyeball, where there are no rods and cone cells, so we see nothing. If the image of an object happens to fall on this gap, we usually don't notice it because the brain fills it with the rest of the visual field. part cannot be used for reading. What this really means is that the neural network starting in the fovea, running from the optic nerve to the visual cortex and extending to other parts of the brain, is trained to be able to read, but the neural network starting in other areas of the retina cannot be used. For reading. All the text we read is scanned by this central area before it enters the visual system.

Short-term memory is not storage - it is not memory and perception. Where the party is processed. More precisely, it is not the temporary repository for information obtained by the sensory system or retrieved from long-term memory.

1. Introduction

For example: In many digital cameras, pressing the shutter button can take photos or videos, depending on which shooting mode is currently selected.

2. Advantages and Disadvantages<. /p>

User interfaces with modes have their advantages, which is why many interactive systems provide modes. Modes allow a device to have more functions than controls: the same control provides different functions in different modes. Interactive systems assign different meanings to the same operations thereby reducing the number of operations the user must learn.

However, a well-known drawback of patterns is that people often make pattern mistakes: they forget what the system is currently doing. This shortcoming is especially obvious in systems that provide poor feedback about which mode you are in. Because of the problem of mode errors, many user interface design guidelines say to either avoid modes or provide strong feedback. Feedback informs the current mode. Human short-term memory is too unreliable for designers to assume that users can remember which mode the system is in without clear, continuous feedback, even if the system's mode switching is determined by the user.

Although Landauer (1986) used the average human learning rate to calculate the amount of information a person can learn in a lifetime, no one has measured or predicted the maximum information storage capacity of the human brain.

The main implication of the characteristics of long-term memory is that people need tools to enhance it.

Since prehistoric times, people have invented various techniques to help them remember things over time: grooved sticks, knotted ropes, mnemonics, oral stories and fireside histories, texts, and scriptures. , books, digital systems, shopping lists, checklists, phone books, diaries, accounting books, oven timers, computers, mobile digital assistants (PDA), online shared calendars, etc.

After millions of years of evolution, the human brain has been "designed" to recognize objects quickly. In contrast, retrieving memories without sensory support must be unimportant for survival because our brains are not at all good at recalling.

Seeing and selecting is easier than recalling and typing. Show users options and let them choose from them, rather than forcing users to recall their options and then tell the computer.

Example:

1. Introduction

The ability to learn from experience has a long evolutionary history. To do this, a creature does not need a cerebral cortex (i.e., a new brain). The old brain and midbrain learn from experience. Even insects, molluscs, and worms don't even have old brains and can learn from experience with just a few clusters of neurons. However, only organisms with cerebral cortexes or similarly functioning brains can learn from the experiences of other organisms. To be aware that you are learning from experience, the cerebral cortex is certainly required.

2. Defects

However, our ability to learn from experience is not perfect for several reasons:

First, for complex situations , such as those situations that involve many variable factors or are affected by many unpredictable external factors, and it is difficult for people to make predictions, or learn and generalize from them. For example:

Second, the experiences gained from our own lives or from our relatives and friends have more influence on us than those we read or hear. For example, we may read or hear reports, customer reviews, and statistics indicating that the Toyota Prius is a good car, but if a sister or uncle had a bad experience with it, we may have a negative opinion of the car. There are negative reviews. We do this because our midbrains believe that family members are more similar to us and therefore more trustworthy than hundreds of other faceless consumers, even though from a rational perspective, statistics are much more reliable.

Third, when people make mistakes, they don’t always learn the right lessons. When they find themselves in a bad situation, they don't remember their recent behavior well enough to connect their current situation to the real cause.

Fourth, the fourth problem people learn from experience is that they often overgeneralize, that is, make one-sided generalizations. For example, many people assume that all crows are black because all crows they have seen are black. There actually are crows that are not black.

1. Problem Solving

Having a cerebral cortex can free organisms from complete reliance on instinctive, reactive, unconscious and skilled behaviors. The cerebral cortex is where conscious reasoning occurs (Monti, Osheron, Martinez, & Parsons, 2007).

However, solving technical problems requires an interest in and training in technology.

2. Calculation

Modern humans evolved from primitive humans 200,000 and 500,000 years ago, but it was not until around 3400 BC that people first arrived in Mesopotamia In Asia (present-day Iraq), numbers and numerical calculations were invented and began to be used in transactions. At that time, the human brain was basically the same as the human brain today. Since the human brain evolved before the advent of numerical computation, it cannot be optimized for computation.

3. Problem solving and calculations both require the help of short-term memory

In contrast, controlled processing, including problem solving and calculations, requires focused attention and uninterrupted awareness. monitoring, and proceeds relatively slowly and sequentially (Schneider & Shiffrin, 1977). This strains the limits of our short-term memory, as all the chunks of information needed to perform a given step compete with each other for scarce attentional resources.

This requires conscious mental effort, as if you were asked to memorize the alphabet in reverse order from M to A.

Omitted

Let users use feeling (experience) rather than calculation.

Cognitive psychologists call the gap between the tools that users want and the operations that the tools can provide the "execution gap" (Norman & Draper, 1986). People who use tools must expend cognitive effort. The cognitive effort converts what he wants to do into the operations that the tool can provide, and vice versa. This cognitive effort draws the person's attention away from the task and puts it on the requirements of the tool. The smaller the gap between what the user wants to do, the less the user has to think about the tool itself and the more they can focus on their task, so the tool can be automated faster.

< p> Constructing an object-operation matrix can give you a visual representation of the complexity of the conceptual model of the interactive system. The larger the matrix, the more concepts the software development team should learn from the conceptual model. Create a product dictionary. In this dictionary, every object, action, and attribute that users can access in the product (including its documentation) has a name and definition. Entries and concepts should correspond one-to-one in the dictionary. It should not be the case that multiple words correspond to one concept or one word corresponds to multiple concepts.

Perceived responsiveness may seem less important than effectiveness, but in fact it is the opposite. p> But a highly responsive interactive system is not necessarily high-performance.

We only consider the 10x level. Similarly, we can get around 10ms, 1s, 10s and 100s levels. Grouping. More than 100s is beyond the scope that most interaction designers need to consider.

The common excuse for not using the busy flag is that the operation is very fast. will end, so there is no need to display the busy indicator. But what is "fast"? What if the operation does not complete quickly every time? What if the user's computer is much slower than the developer's computer or if the operation is not optimized? What if the data to be read is temporarily locked? What if the operation requires access to a network service and the network is congested or disconnected?

Should the software prevent the user from continuing to the next step during operation? The operation displays a busy flag, even if the operation can usually be completed quickly (for example, within 0.1s), this flag may be very useful for users to wait dynamically in case the operation is blocked or crashes. Logos are more user-friendly than static ones because they show the system is working rather than crashing or hanging waiting for a network connection or data to be unlocked. Of course, busy dynamic identifiers should be synchronized with the actual computation they represent.

When displaying the percentage of an operation that has been completed, start from 1% instead of 0%.

The display of progress should be gentle and linear rather than unstable and explosive.