Are UX Laws really effective?
With the help of User Studies, questioning the validity of the current UX Laws and their scope of usage.
Background
An exploratory study done in collaboration with Yamini Gupta.
In this study we tried revisiting conclusions made by UX Laws, testing out their effectiveness and is their application during the design process really that useful. Before we started this project, we felt UX Laws missed out certain key areas that a designer needs to focus on during the product development phase, an example could be of how it’s equally important to not just consider Users of the product as the only stakeholder, considering Business goals, development constraints and what competition has to offer is equally important, many of times these things don’t overlap with the conclusions made by the Laws of UX.
Study Participants
One of the major constraints we felt was the shortage of participants that took part in our study, this being an exploratory project we still went ahead with the limited resources we had. Our test participants in total included 4 college students. All of the participants are of Indian ethnicity with no apparent difference inability, and all are frequent smartphone users.
Aim of the study
- To shed importance on the fact that it’s important to consider multiple/various factors while designing anything.
- To tell that context could sometimes play a vital role in how we process information than the information itself. And that’s the point we aim to highlight: that the present UX laws strictly apply to the information but don’t include the context of the information.
- To highlight how important it is to always consider multiple stakeholders, UX laws do consider users, but it’s just that. They don’t consider the fact that there are multiple stakeholders that collectively contribute and involve in product/service development/usage.
- When we want to run a website we need both front and backend, missing of anyone wouldn’t let our website work. Same is when we consider the goal of ours to create an experience for the user. When we talk of experience of digital products it’s about how visual experience is working in together with information experience and personal objective experience. UX laws are primarily focused towards the visual experience but misses out that in multiple cases prioritising visual experience over others could actually backfire.
Constraints
- Study /participant pool size
- Limited context and use cases
- Targeted limited UX laws
- Developments over time, constantly evolving Human behaviour and Design practices.
Laws Scrutinized
Though there are multiple UX Laws, we scrutinized only 6 of them. The selection was on basis of how popular they already are, the initial thoughts we personally had regarding them, and their scope.
The laws we tested out include Jacob’s law, Aesthetic Usability Effect, Zeigarnik Effect, Serial Position Effect, Miller’s Law, Law of Similarity.
Aesthetic Usability Effect
Current belief
Aesthetic Usability Effect states the positive relationship aesthetics and perceived usability have. If a user finds a design aesthetic, they will be more forgiving or lenient of its usability. This is because the user forms a positive emotional relationship due to the aesthetic pleasantness of the design, which causes them to actually perceive it better than it is.
For example, in a usability testing user research conducted by a design agency, one user encountered many issues while shopping on the FitBit site, ranging from minor annoyances in the interaction design to serious flaws in the navigation. She was able to complete her task, but with difficulty. However, in a post-task questionnaire, she rated the site very highly in ease of use. “It’s the colours they used,” she said. “Looks like the ocean, it’s calm. Very good photographs.” The positive emotional response caused by the aesthetic appeal of the site helped mask its usability issues.
Why we challenged
Over the course of time, the interpretation of Aesthetic Usability Effect has become highly distorted. It doesn’t take into account the subjectivity of aesthetics. All designs have started following a set pattern of aesthetics, hence mirroring Jacob’s Law. This is so because it automatically makes the user feel that the design is usable as well.
Since a set definition of aesthetics in design has started to be manifested, no experimentation around the context and its dynamic are being done. For example, a design may seem more aesthetic objectively, and hence more usable. However, if a context is attached to it, the user may no longer feel that the design is aesthetic.
More so, an increased focus on aesthetics leads to a bad User Experience. Because even if the user is more lenient to the usability on a short run due to the aesthetics, a repetitive annoyance in regards with usability will start to creep in. This will hence damage the emotional relationship that the user has built with the design, so much so that the user might even start considering it non-aesthetic.
How we challenged
There are 2 definitions that need to be noted here:
- Perceived usability: The usability of a product based on your first impressions of it, which may or may not be involuntarily biased by an unknown previous experience. It is the usability that the user perceives the design will have based on a setting/context that they’re imagining it in.
- Contextual usability: The usability of a product based on the context that has been provided to the user. The user now needs to imagine the design in that specific context and then evaluate whether the design is usable or not.
Two sets of products were chosen: 2 different chairs, and 2 different pens. In each set, the two products had completely different aesthetics in terms of texture, material, appearance, etc. However, their cost was the same so that there is not much evident disparity in the designs.
The users had to rate the objects on a scale of 1–5 for aesthetics and usability, with 1 being the lowest and 5 being the highest. Once they had rated it, a situation was presented to them where they had to choose which product they think would be more usable. It was then noted whether their answer changed from the previous answer or not after attaching the context.
Result
Gathering results from 4 different users, the perceived usability for the 2 designs in a set was more or less the same, while the aesthetics varied greatly. By the aesthetic usability effect, the usability should’ve correlated with the aesthetics more. However, all the users treated aesthetics and usability as two independent terms that have little to no correlation between them.
In the graphs seen above, it can be seen that the aesthetics of design-1 in both the sets is rated higher than that of design-2 by all the four participants. However, the usability of design-2 is either greater, or equal to that of design.
Out of the 8 scenarios (4 people x 2 sets of products), in 7 scenarios the aesthetics and the contextual usability did not overlap. While the users found product 1 more aesthetic, they found product 2 more usable when provided with the context.
Jacob’s Law
Current belief
Jakob’s Law recommends the use of familiar patterns in design in order to facilitate user experience because users prefer it when a site works in the same way as all the other sites they already know.
For example, people are familiar with the format of Horizontal Scroll when there are presented with data such as ‘Friend’s stories.’ We have seen a very similar design in apps like Instagram, Snapchat, Facebook, and now LinkedIn. And it’s undoubtedly true to an extent that people are indeed familiar, and hence comfortable, with the design styles they have been already exposed to. However, this is also a drawback as this does not pave way for creativity and innovation.
Why we challenged
Continuing the above example of Horizontal scroll in Stories, Twitter executed the same idea in a slightly different format. Instead of horizontally scrolling the stories of the same user, you do the same navigation vertically. For some, it proved to be of better Usability. This was a risk that Twitter took, but many companies do not.
Similarly, Apple took the risk with IOS 7 or Google with Material Design, and these bets proved to be right. Jacob’s Law biggest drawback is how it makes development remain in a loop by not promoting risk-taking culture, meaning it doesn’t support out-of-the-box ideas and limits the design innovations.
A more feasible/effective solution cannot be devised using this logic since the user is not habitual to it yet. Here, compliance is being prioritized over temporary discomfort. In saturated markets as well, following Jacob’s law would actually result in limited reach since for young products it’s always important to highlight themselves.
How we challenged
The study was divided into 2 parts
- First Test was targeted to prove how usability preference and habits differ from person to person. The difference could be based on their existing usage, their age, language etc.
The first test includes a Usability score test(score to be given by the users only). Users will be provided with 2 banking apps with a different layout. We would be given them the phone with the banking application open(at the home screen).
- This is the second test where we explore the context of the designed stuff and the visual sense involved.
In this, two portfolio websites of two visual designers is selected. One design is very conventional and minimal while the other is very different and swanky. We provide them to our participants, who are aspiring designers, and ask them which one they like more.
Results
The test results overlapped to what we believed that Usability preference changes from Individual to Individual and how Jacob’s Law hinders creativity.
When we presented them with the layouts of two different banking apps, we found that the layout that has a bottom navigation bar was prefered by young test subjects while the one with simple icons placed at the centre was preferred by older test subjects. The reason could be existing habits or apps used by our test subjects, social media apps like Facebook or Instagram is general have a bottom navigation bar, highly used by our your test subjects could be the reason for their preference.
While in the portfolio review round results were slightly different than anticipated. while all of the test participants appreciated the creativity of the portfolio 1 but at the same time complained about its usability. 2 of the 3 test participants hence went with the 2nd portfolio as their preference citing simpler and more usable design. Though all of them agreed that Portfolio 1 is more innovative and different.
Serial Position Effect
Current Belief
According to the serial position effect, the user would always remember the first and the last items of the list more than other items.
Why we challenged it
We felt that this hypothesis needed some edits as it completely misses out on multiple crucial factors that work simultaneously when the user is grasping the data from the list.
Such crucial factors could be:
The length of the list; lists that user may consider lengthy and not worthy enough to look at it completely. In such a situation, the user would never reach or focus on the end result/item.
How the information is classified; information containing only text would be perceived in a different way, information in form of search results or general would be perceived differently.
Other than that, the serial position effect misses out the factor of emotional or any other connection with the content. For example: in a list of movies a person is more likely to remember the names of the movies that they have already watched and are on the List.
How we Challenged it
At first, we provided our participants a lengthy search results [(a) A Wikipedia Page, (b) A Long list of Movies] and told them to look at them. Later after a specified interval of time, we tried to grasp how much and what all things they remember in the recall session.
The second step included a list of names. Different names were chosen with specific purposes; for example the list contained some Indian names, some Russian or some Chinese, and the placement of these names was also thought of in advance.
In the third step, we tried to analyze whether there is a correlation between serial position effect and decision making by presenting them with a payment plan screen.
Results
The results overlapped to what our expectations were.
In this case, the major causes of why people weren’t able to recall the items placed at the end of the list were because of their overall interest and time provided to them to look at all of the items. People tend to skip parts and focus only on the certain section (most of the time, the top sections) of the list provided if it is too long or it doesn’t hold much value to them.
Results also showcase how we remember that information: it's based on something we could relate to or find easy to process. For example, when names were provided to them as a part of our study, they were able to recall those names more which they can pronounce or they find interesting enough that they devoted more time on them.
Concludingly, we found Serial position effect to be partially true. People tend to focus on the first item on the list, but certainly not on the list unless the list is too short and we are creating an artificial environment for them where we are focusing on the fact that they look at the list and remember the items it consists of, a situation that is unrealistic.
Miller’s Law
The number of objects an average person can hold in working memory is about seven, also known as The Magical Number Seven, Plus or Minus Two.
— Miller’s Law
This is critical because humans are constantly performing tasks, and trying to juggle various stimuli in the mind when doing so.
Why we challenged
It appears to us that magical Seven only applies to Uni-dimensional stimulus in absolute judgment and items in an immediate memory span, and in each case, they behave differently.
Even though Miller’s Law mentions of the concept of Chunking for better efficiency of our Short-term channel memory, there are many more crucial factors being missed out, many of which factors are very similar to factors we believed Serial Position effect misses out.
For example, there needs to be more mentioning of the classification of data presented to users. Our brains’ receptiveness to images could be different to text. Much like serial position effect it also misses out the possible connect data could have with the user/participant. Images they are already aware of compared to images they are not would yield different results. It also misses out humane factors such as age, present Intellect level, etc. There’s no denying that as we age our working memory also loses its efficiency.
No way out from Miller’s Law bottleneck, UX Designers need more ways they could present data to users, apart from chunking. These include studying the time period for which the user looked at the data or the factor of Gamification in the consumption of data.
How we challenged it
A focus group testing in parts will be needed to be done. Users will have to be exposed to different permutations and combinations of numbers, images, names, random words. In the first case, there will be two lists of number and test cases are exposed to it, the difference is, we have slightly gamified the experience of learning or looking at the second list
2nd test of Miller’s Law was very similar to the 2nd test of the Serial position effect, two lists were provided to test subjects, the first list consisted of names in the textual format while the second list contains images, this test gave us an idea the difference is created in remembrance based on the type of content.
Results
Results of the first test highlighted two major things.
Firstly, working and short-term channel memory vary from person to person and some are able to remember more, some less. This is could also be a result of how they process the information, like naturally making chunks.
The second thing that was highlighted was that even if the results varied they weren’t too far from the number proposed by Miller’s Law that is 7+-2.
Gamification does make a change but not substantial, lack of test subjects also questions whether on a bigger scale it does affect the efficiency of our brain in the processing of information. Substantial changes to make a claim were only observed in 1 out of 3 participants. We observed, more than gamification it is the bait of incentives that motivates people to be more aware/alert.
The second point showed how the type of content we are consuming is responsible for the way we process it, but again the changes were not too substantial. It showed how we are able to remember images or visuals more than text. Also if the content is interlinked, the mind is easily able to remember all the items by thinking of that link. For example, different names of movies that are all part of the same franchise. You may remember more but even if you do so the final bits of information our brain can hold is not too far from the number proposed by Miller’s Law.
Zeigarnic Effect
Named after Soviet psychologist Bluma Zeigarnik, in psychology, the Zeigarnik effect occurs when an activity that has been interrupted may be more readily recalled. It postulates that people remember unfinished or interrupted tasks better than completed tasks.
Why we challenged it
We need to define the task. Tasks could be small interactions like navigating to a specific screen or it could be completing a video editing project you wanted to.
Also, the value of the task for the user is directly proportional to the Factor of Satisfaction. If a person has a goal that they have been working towards, and then they achieve this goal, they’ll feel a sense of satisfaction. This satisfaction makes them feel proud, and when they feel proud they tend to remember that feeling more. This is hence a factor that has been overlooked in the law.
Other than that, the law is missing out the point of how important it is to alert customers that they have not completed a certain specific task and they should try to complete it. It’s important that the user is notified again and again the steps or the process he missed out or didn’t complete. If we don’t focus on such an alerting system then there’s a high possibility that users would skip it.
How we challenged it
The study was divided into three parts.
The first includes a Short five-question test including questions of Quantitative aptitude, test participants were given 10 minutes for the exam, afterwards, some pre-specified questions were asked to them regarding the questions they just attempted.
The second part included asking general questions to them which were decided beforehand and were targeting information that could help us analyse the correlations such as a personal interest in the task and how it affects claims by Zeigarnic effect.
The third and the last part was a continuation to part 2 only, we gave our test subjects a long paragraph to read, while they were reading we interrupted them at 4 places, 1 of these 4 places in the paragraph where they were interrupted included content they could relate to more. Afterwards, we asked them to recall all of the places they were interrupted for analysis.
Results
Results were very similar to what we expected and did highlight the cause of concerns with the Zeigarnic effect.
From the first test of Quantitative Aptitude questions, we got the insight that the participants remembered questions which they were able to solve. Although, 2 of them also did remember the questions where they were stuck, which was further validified by step 2 of the tests conducted. In the second step, we asked them about their gaming IDs. Since that was something important to them, they remembered the question, even though they weren’t able to remember the answer in the first place.
When we analyzed the last step, all of the participants remembered 1 point out of 4 where they were stopped: it was the point that catered to their interests more.
The results clearly showcase how our satisfaction after completing a task is directly proportional to the chances of us remembering the whole process. Also, the probability we will remember an interrupted task is again dependent only on how important that task it is for us and how we are revisiting it.
Law of Similarity
Under the Law of Similarity, the human eye tends to perceive similar elements in a design as a complete picture, shape, or group, even if those elements are separated. These similar elements could be anything like shape, size, type, colour, or proximity.
For example, in the instance of a magazine page which has body text, an image, and a caption, the body text and the caption are similar-looking since they’re both texts and yet they’re not grouped together. In fact, the image and the caption are grouped together.
Why we challenged
In a design wherein too many elements need to be placed, with slight similarities and dissimilarities among them, it becomes imperative for us to judge which elements will be grouped together by the viewer. Inability to judge this correctly can lead to excess confusion, and hence a bad User Experience.
For example, in a website that offers a multitude of services, it is important that the placement of the CTA buttons is well planned and that the respective functions of the buttons are well communicated. Especially in E-commerce/SaaS websites, lots of CTA buttons are placed in a haphazard manner, and their descriptions are not aligned well. The disproportionate proximity and similarity of elements lead to confusion.
How we challenged
Users were presented with a total of 20 permutations and combinations: both involving text and shapes. These text blocks and shapes were altered in terms of position and appearance using elements such as size, colour, proximity, and type. The users were then asked to group the elements together, and their responses were noted down.
Results
The users grouped the elements based on type, proximity, size, and colour in the given respective order. Two sets were worked with: text blocks and shapes, to see whether there is any difference in the results produced. However, the answers of the respondents mirrored in almost every situation. Whenever the elements were grouped according to type and colour, the participants favoured type. The same case was replicated with proximity and size as well.
Conclusion
Largely results of the study overlapped with our expectations, highlighting how UX Laws are effective and useful for designers only in certain curated conditions. Many times situations are different involving multiple stakeholders, and UX Laws try to bring all these stakeholders and all of the users under one umbrella.
We wish to further continue on this with a much larger pool of participants for more validation of the results we got regarding each individual UX Law, primarily highlighting the need for more Diversification and edits in the existing definition of the Laws.
:D
Aditya Bansal
Interaction Design student at Delhi Technological University
Reach out to me at aditban@gmail.com or at adityaban.com