How our expectations determine how we process information

Both, top-down and bottom-up processing are the two famous approaches on how we interpret information in cognitive psychology and as humans, we do both.

Please read the sentence below:

Now read again.

Have you skipped over the extra “the” or did you read the line for what it is?
Let’s be honest: We almost all skipped over the extra „the“ and the reason why we do this is that we do use top-down processing, which is one of two ways we are interpreting information about the world.

Top-down or conceptually-driven processing simply means that the processing is hugely influenced by the individual’s expectations which are based on previous knowledge rather than by the stimulus itself. The latter is called bottom-up or data-driven processing which means interpreting the stimulus solely for what it is.
There are many examples out there that we use top-down processing in combination with bottom-up processing – for example, read the following sentence:

„It dseno’t mtaetr in waht oerdr the ltteres in a wrod are, the olny iproamtnt tihng is taht the frsit and lsat ltteer be in the rghit pclae“.

Fun fact, which also proves we are using top-down processing: If your English is weak (or you’re not a native speaker), you might have a much harder time reading this sentence correctly as someone who speaks English fluidly: she could easily read this paragraph without hesitation. If you are German you might fluidly read a similar example in German:

„Gmäeß eneir Sutide eneir elgnihcesn Uvinisterät ist es nchit witihcg, in wlecehr Rneflogheie die Bstachuebn in eneim Wrot snid, das ezniige, was wcthiig ist, ist, dass der estre und der leztte Bstabchue an der ritihcegn Pstoiion snid“

So we see, how we process information is determined on previous knowledge and expectations, and how we process information is also guiding where our attention goes (there are several other ways such as priming and other biases but we’ll omit that by now). For example, there is this famous young woman/old lady illusion which is one of the best examples of perceptual expectancy:

Whether we see an old woman or the young lady is due to our interindividual top-down processing differences.
We can literally say that we perceive what we expect and know —if there is no prior knowledge of something, the tendency to overlook details is rather high because we have no (strong) association with something meaningful to us.

How might this affect Interface Design?
As we did see, perception and information processing is not objective which means simply just because there are certain elements placed somewhere people will see and use those elements.
The way we look for information is not only feature driven (or data-driven which equals bottom-up processing) but also context-driven or expectation-driven which equals top-down processing.

And to make sure that both approaches are met and people can find the things they are looking for it’s important to gain average knowledge in form of basic user research of the people who are using the product, site or app, and their goals and expectations.

Stop treating (usability) tests as a substitute for explorative user/design research

Usability tests / evaluating systems are task-based whereas user research is rather explorative – while researching we are exploring the goals, needs and also fears of the people who will use something and gain knowledge about them which will inform the basis of any tool.

Cognitive models as a substitute for quantitative usability tests?

„Cognitive models can serve as a substitute for (quantitative) user tests. User models built with ACT-R can simulate the interaction with a certain task. Cognitive modeling has two advantages over real user tests; first of all no human participants are needed when good and evaluated models exist and second, important information about underlying cognitive processes can be discovered. Implications from these findings can then be used in designing further applications.“

Russwinkel, N., & Prezenski, S. (2014). ACT-R meets usability. Or why cognitive modeling is a useful tool to evaluate the usability of smartphone applications. Paper presented at Cognitive 2014: The Sixth International Conference on Advanced Cognitive Technologies and Application, Venice (pp. 62-65).

Huh. So many questions. I guess a computational model generally de-emphasizes or even neglects human emotional and affective factors as any other feelings such as stress, tiredness and their implications on motivational factors. (I think I read somewhere that ACT-R has built in some motivational component?)

I don’t think computational models could substitute humans when it comes to usability evaluation even when the evaluation is only based on efficiency data, but it is definitely a very interesting approach which caught my attention and now I’m curious :)

John McCarthy on AI

Stumbled upon this goodie while researching on cognitive architectures for a course.

„(…) a machine isn’t the sum of its parts, if somebody took a car apart and gave you a heap of the parts that wouldn’t be a car – they have to be connected in a specified way and interacting in a specified way, and so, if you want to say that the mind is a structure composed of parts interacting in a specialized way I would agree with that, but it isn’t just a heap of them“

October edition of Ladies that UX Berlin is coming along!

Ladies that  UX Berlin October 2017
Our next Ladies that UX Berlin meetup will be Tuesday October 24 hosted by Mozilla Berlin. The location is GSG-Hof Schlesische Straße 27, 10997 Berlin; Gebäude 3, 4.OG – There will also be some signs in the courtyard and staircase. Event opens doors at 6:45 and there will be two talks this time: One by Marja Annecke (SinnerSchrader Swipe) – she will be talking about CX & Branding and one by Sabrina Mach (ThoughtWorks) about how to use acting techniques to understand & predict human behavior.

A big thank you to Mozilla Berlin for hosting and sponsoring this month’s event.

Make sure to RSVP here, if you are close by! See you there!

Contact me if you have a topic you’re passionate about and would like to present at an upcoming event or would be interested in hosting one of our upcoming events

On cognitive dissonance and rewards

„We try to reduce the dissonance between how we think we should act and how we actually act by changing one or the other“

This is interesting because it is contrary to incentive / economic theories which claim that the higher the reward will be the more likely people will change their mind – but only if there is a mismatch between my internal attitudes, values an or core beliefs and how I actually act.

The video shows excerpts of a classic experiment in social psychology conducted by Leon Festinger and James M Carlsmith in 1959 which is called „Cognitive Consequences of Forced Compliance“. Forced compliance is very closely related to the theory of cognitive dissonance which states that there will be the mental discomfort (psychological stress) experienced by a person who simultaneously holds two or more contradictory beliefs, ideas, or values. This, in turn, is related to one of the main principles in Gestalt Theory: The principle of good form.

GDPR Cookie Consent mit Real Cookie Banner