In one of my previous articles I wrote about ways to design for human attention. Attention is closely linked with working and short-term memory. However, this article is going to be focused on the entire human memory system. We have two main types of memory: short-term memory (STM) and long-term memory (LTM). In computer science terms, you might imagine short-term memory as RAM and long-term as the HDD. I will try to describe how they work, how we can design interfaces that eliminate confusion and how to lower the cognitive effort users need to make. All of this should result in a more human-centered user experience.
“The human brain is not optimised for the abstract thinking and data memorisation that websites often demand. Many usability guidelines are dictated by cognitive limitations.” – Jakob Nielsen, Nielsen-Norman Group
LTM is the memory for data storage. Memories saved in our mind are basically clusters of neurons linked together. To activate a memory, we have to reactivate the same pattern of neurons that was stimulated when the memory was formed. Most memories from our STM are forgotten. This is perhaps a good thing. If we didn’t automatically forget the huge volumes of information we are exposed to on a daily basis, we might get overloaded with information, as the processing of these volumes of data would soon become impossible.
To transfer perceptions from STM into LTM, conscious effort has to be made. Repeating particular pieces of information results in data transfer into LTM. This is the reason we revise before exams. The brain has to form a neural memory pattern of small pieces of connected data. Afterwards, the pattern (memory) can get activated by:
- Sensory input from our environment
- Stimulation from other parts of the brain
This results in the successful retrieval of the memory. The more times a pattern gets reactivated, the stronger it gets — it is easier to reactivate.
Facts about LTM:
- Human brain’s capacity has not yet been confidently established. For now, it seems as if the capacity was flexible and sufficiently large to keep surprising us!
- People remember experiences connected with emotions. For example: You remember the first time you drove a car but not the 156th.
- The data stored there is not a high-resolution representation of each particular experience. In computer science terms, we might say that the brain uses a very efficient compression method and lets go of a large portion of the information. Different memories are also stored at a different level of precision and detail.
Psychologists explain STM as a combination of perception, attention and retrieval from LTM. If one of our senses is stimulated, it creates a residual perception available as a possible input for our attention. If we decide to grant attention to this specific input, it is allowed to enter our working STM.
Long-term memories reactivated through retrieval are also available as input. The main component of our short-term memory is a combination of a small subset of our sensory inputs and temporarily activated long-term memories that we are aware of right now.
Facts about short-term memory
- According to George Miller, STM can only hold about 7 (+-2) chunks of information (Miller’s Law).
- A brief period of time during which the STM can hold information is between 15 and 30 seconds. After that, the information is either committed to LTM or forgotten.
Memory in interface design
As designers, we should strive to design for the STM. The memory load is lighter and the interaction is faster and more error-free. If a user has to recognise or recall something from the LTM, that takes time, cognitive load and increases error occurrence probability.
Hints for the user
Distinguish visited links with different colours. Users don’t have to remember which links they have already visited.
Many elements at once
Don’t present users with many elements in one interface. It is not necessary to restrain oneself to 7 elements as Miller’s Law would suggest, but the elements need to be distinguished enough — perhaps by using visual hierarchy. Otherwise the interface will be a burden to the user’s STM. Choosing the correct way to continue in the flow will become more time-consuming. If you need to use a lot of information, divide the journey into more steps.
Consistent design system
Create a set of standards throughout the product. Adopt widely applied and tested design patterns. Once you do that, users won’t have to constantly learn and remember new patterns, but they can use the knowledge they already have. If a confirmation button in the product is green, all other confirmation buttons should be green as well, easing the user’s cognitive burden.
Companies have already discovered the power of design systems and are applying it throughout their portfolios of products.
Users will remember your interface more clearly if they feel emotions during it’s use. Emotions can be triggered in many ways. Aesthetically pleasing visuals, fun communication or simple gratification for solving a problem. Once this is achieved, people will gladly return and use your product again. On the other hand, users can also happen to feel negative emotions, which will result in forgetting the details of the interface, but remembering your design as a negative experience. Then, churn happens.
The Self-Generation Effect
Information is better remembered when it is self-generated — not just passively consumed. If a user can customise the layout of the interface, it is easier for him to become familiar with it and remember the details.
Always put reasonable demands on password creation. Setting password creation standards too high might seem as an effective measure to increase the security of the system, but it will most likely result in making the system more vulnerable. Average users will be forced to write down the password and create another burden for their LTM, as they have to remember where exactly they wrote it down.
In the past, people had to use command lines as the interface to get to their goals. This is an example of design that creates a big load on LTM and a high error occurrence probability. Luckily, with the invention of GUI (graphic user interface), designers now have a lot more freedom to design experiences. It is much easier to focus on using STM and on easing the cognitive load. It almost seems like users don’t even have to think while using today’s technology. And the future looks promising. With the rise of virtual and augmented reality, we will be able to create products that we can interact with easier and with less effort than ever before.
This article was originally published on Martin’s Medium page.