EuroSTAR: The Testers Toolbox – Seven Powerful Cognitive Techniques

Graham Thomas introduced seven powerful cognitive techniques for the tester’s toolbox at the EuroSTAR conference.

Graham presented the techniques in a workshop.

The first technique Graham showed was Gall-Peters Projection. Graham showed two projections of the world based upon Gall and Peters. Gall made Europe the center of the world and made it rather larger than it actually is, while Peters focused on making other countries larger in the projection to make it more accurate. Graham continued with different representations from the world’s past about how the world looks. He described word-maps from tag clouds or tub maps, or critical path networks. Graham collected responses of models and maps from the audience. Among them were test results, and test reports which leave out essential detail, as well as to not believe models, or rather use models in the right way.

The second technique was Popper’s Theory of Testability. Graham gave the definition of the theory:

Falsifiability or refutability is the logical possibility that an assertion can be shown false by an observation or a physical experiment. That something is “falsifiable” does not mean it is false; rather, that if it is false, then this can be shown by observation or experiment. The term “testability” is related but more specific; it means that an assertion can be falsified through experimentation alone.

Graham also referred to black swans. Graham explained how to use the Theory. Whenever you see All, or Every, or Never you only need one counter-example to falsify the claim. You can use this understand success and failure criteria. Last, if you only have a short amount of time, you can use the theory to target your testing.

From the second exercise the audience said that Popper’s Theory can be used to challenge assumptions. Positive verification is not enough, we need testing for negatives, too. As testers we should not believe what we see or think. We have to check our requirements first. Getting evidence on multiple levels is worth for the tester’s mindset.

Graham continued with his third technique, which is mind control. He asked the audience questions, and had made up some predictions in his slidedeck. He asked the audience to think of a vegetable, flower, color, tool, a number between 1 & 10, and a software testing technique.

Continuing on why this is important, is that when we ask questions some people may be hard-wired to give certain answers. If we ask questions, we have to listen to the answer, Is it actually the answer to your question? It could be GroupThink, the answer that everyone in your culture or company has. Maybe it is just the lowest common denominator., like boundary value analysis. This leaves the question open, on how we should ask questions.

Graham showed Bloom’s taxonomy of educational objectives. The categories there were knowledge, comprehension, application, analysis, synthesis, and evaluation.

Getting back to the audience, the collection of things around mind control were confirmation bias, and buzzwords as in the GroupThink camp. Authority might have something to do with mind control. Impossible to answer questions for testing like “Are we finished testing, yet?”. Incongruent body language plays also a role in this regard. If the words, and the body language don’t agree, then this might control the minds of the others in the meeting room. Standards are controlling other peoples mind as well.

Graham continued on the Stroop Effect. He had decline the audience the color of the letters appearing on the video screen. He showed that our eyes are more powerful than our mouth. Color is basically hard-wired into our brains. We have been conditioned through evolution to respond to some colors differently, for example red means danger in most contexts. On color blindness and visual impairment, Graham said that 8% of the males are affected, and 0.5% of the females.

The audience replied to the Stroop Effect with things like usability in terms of color information. Accessibility is also a matter. Color bias can be unlearned to some extent. Context for color-bias matters. And color perception can change with experience and culture.

Graham continued with the cognitive techniques with the Necker Cube. HE showed an animation with two dots on a cube switching back and forth. At some point the dots switched from appearing to switch left to right rather than from top to down. The switch happened for some people in the audience, for others it didn’t. The switch happened for people only in their brain, Graham explained. It is therefore important to get what others see. It is important to understand that others may see different things than ourselves. We don’t see things as they are, we see things as we are, Graham answered the question on how we see things.

Asking the audience what others see, and what we may use from the Necker Cube in our software projects, the ideas were that the business may have a different view of the requirements. One thing was to consider what others think about Exploratory Testing. A picture says more than one thousand bugs. When you see something, ask what else could this be. The word “unless” is very useful in this regard. Paraphrasing is one aspect to overcome the Necker Cube. Pairing on testing can also help with different aspects of the product.

Graham’s sixth cognitive technique was the Spinning Dancer. Graham explained that you get a different view when you look at the whole to that when you look at the individual parts. We shouldn’t assume that our view is the only view. Testers should think of this as a process. People in the different parts of the process see the process working differently. You need to understand the individual viewpoints. For example the defect management system on a large testing program may be perceived as a bottleneck though this might be a perception for just the testers.

Asking the audience whether they see the individual parts or the whole, the replies included that every layer gives another view of the whole. Taking a layered view might be something to try out. Sometimes things need to be de-constructed in order to understand what is going on. One response challenged whether the dancer was dancing, or the surrounding was changing rather than the dancer. One proposed a Yes/No and Don’t know tristate boolean approach to questions. Objects can also move while they are still.

The seventh and last cognitive technique Graham defined E-prime:

E-Prime(short for English-Prime, sometimes spelled E′) is a form of the English language in which the verb to be in all its forms does not exist. Thus, E-Prime does not contain the words “be”, “is”, “am”, “are”, “was”, “were”, “been” and “being”, nor does it contain their contractions “‘m”, “‘s”, and “‘re”. E-Prime therefore uses alternate means to express most statements which use the passive voice, thus encouraging writers and speakers to clearly state an action’s agent.

Graham explained that E-Prime is about how to communicate experience rather than own judgment. He referred to the online E-Prime tool. We tried out “The test strategy is good.” that the tool didn’t like at the word “is”. “It seems good” was something the tool accepted as being in E-Prime. “The strategy makes me happy” is another example which works well. “Compared to other test strategies this one gives some lead” another one. Play around with the tool on your own to find out more about it, and maybe see the pattern.

Graham showed that it takes some time to get your head around E-Prime, but it helps you to write better English, he claimed. Instead of saying “Your Report is not very good” in E-Prime I say “I do not like your report”. Rather than “It needs to be improved” in E-Prime I will say “You need more accurate data”.

E-Prime enforces a personal frame of reference, the audience said. It helps us testers to argue better. Our own perception can be explained to the other person that I am talking to. It helps to produce constructive criticism.

Graham concluded with “To find fault is easy, to do better may be difficult”. This citation is from the Graak/Roman historian from 46 until 120. What did they know 2000 years ago, that we have lost knowledge about during that time.

  • Print
  • Digg
  • StumbleUpon
  • del.icio.us
  • Facebook
  • Twitter
  • LinkedIn
  • Google Bookmarks

Leave a Reply

Your email address will not be published. Required fields are marked *