Disfluency and how ugly fonts help people remember – on 18th January, 2011

How reducing cognitive load is not always beneficial to users, and what this might mean for designers


  • Fluency studies demonstrate how we have the potential to make big improvements in the performance of our communications
  • It has been found that students remembered and retained information better when it was presented in a disfluent (ugly) font.
  • You may be able to increase users memory retention of detailed product information by presenting it dis-fluently.
  • Could dis-fluency affect design too?


I heard a fascinating interview on the Radio4 the other morning. On it the blogger and researcher Jonah Lehrer was describing a recent study into memory retention.

Here’s the current thinking –  “Many education researchers and practitioners believe that reducing extraneous cognitive load is always beneficial for the learner”. Therefore it has been widely accepted that simpler, more readable fonts – should reduce congnitive burden and therefore aid memory.

Well that seems very logical.

But Jonah had an aching worry – that – as it becomes easier to consume information, the less information will be retained. In his blog post “The future of reading” he describes his thoughts.

Scientific corroboration

In a follow up post “The benefit of ugly fonts” he reports on a recent scientific study – in which students were given supplementary material in a variety of disfluent fonts (Monotype Corsiva, Comic Sans Italicized and Haettenshweiler) while the other group was taught with the usual mixture of Helvetica and Arial. The font size remained the same.

After several weeks of instruction, the students were then tested on their retention of the material. In every class except chemistry, the students in the disfluent condition performed significantly better than those in the control-fluent condition. Here are the scientists:

This study demonstrated that student retention of material across a wide range of subjects (science and humanities classes) and difficulty levels (regular, Honors and Advanced Placement) can be significantly improved in naturalistic settings by presenting reading material in a format that is slightly harder to read….

The potential for improving educational practices through cognitive interventions is immense. If a simple change of font can significantly increase student performance, one can only imagine the number of beneficial cognitive interventions waiting to be discovered. Fluency demonstrates how we have the potential to make big improvements in the performance of our students and education system as a whole.

Jonah went on to talk about the impications this might have on graphic design and therefore web and user experience design – which immediately woke me up on my drive through the rain into work.

So what are the implications?

On the face of it this could mean that all visual communication where we want people to remember our message should be in a disfluent (ugly) font.

Hmm – are we jumping the gun here?

I think there is one big difference between the control group and the type of communications marketers and advertisers are involved in.

Within the school setting people are reading material they ‘want and desire to read and retain’, and it seems, from this study conclusive that in these circumstances retention can be increased.

But in the world of marketing communications we are often trying to ‘capture’ peoples attention – they don’t have to read our messages – and are probably distracted with a load of different cognitive burdens. So maybe putting our main marketing messages in disfluent fonts won’t help increase response rates. The key for this type of communication is simply ‘communicating the message at all’ and therefore the simpler the message the more likely it is to communicate.

The design and use of fonts should all act in a way as to communicate the message as a whole. A disfluent font might be appropriate or inappropriate for any given communication.

An example

An organic farm shop’s might produce some labelling for a product. The use of  flowing fonts with ligatures and complex shapes will help communicate the nature of the product. In fact, if the packaging was in minimalist Helvetica, and poorly executed I might have second thoughts on the quality or authenticity of the product.

The opposite might be said for a company selling super modern minimalist furniture.

But wait a minute there IS something to this. Maybe its when we come to presenting detailed information (like product information and technical information) that we can start considering more awkward and disfluent fonts. After all at this point the person has made a commitment to read the information – just like the students – and is therefore likely to be more willing to retain the information and recall it later.


When marketers and designs are trying to capture a person’s attention, or communicate a very important message , the focus should be on the message, which the use of typography should support.

When a user is reading though longer pieces of material – and have made a commitment to read (think product description and details), then maybe we should consider presenting the information in a more disfluent font.

Parting thoughts

As the scientists say – “The potential for improving educational practices through cognitive interventions is immense”. Could fluency affect both design as well as type? On the face of it as a designer I would say no. Design is all about balance, harmony and rhythm – is that fluent or disfluent?

As a user experience designer I spend every moment of my working life trying to reduce cognitive burden by making information more accessible and ‘easier’ to consume – but if disfluency applied to design too then is this could be turned on its head?

  1. Interesting, but I suggest a slightly different explanation leading to slightly different conclusions.

    I suggest that the increased retention isn’t a result of the fonts being ugly. It’s a result of the font’s being unfamiliar. I have read that the mind is continually trying to guess a word or sentence meaning. So an unfamiliar font (possibly, but not necessarily ugly) makes the interpretation more difficult and slows reading. This increases retention on a cellular memory level by resulting in a longer period of time the synapses are activated in comprehending the letter forms, word forms and sentence meaning. Because the mind processes in parallel, every second the optics recognition part of the brain is trying to figure out what letter or word is represented by the shapes, is also a second allowed for the parts of the brain dedicated to remembering and associating to do their work. Have you ever strained to read a traffic sign in the rain on a highway and if so you were probably more likely to remember the sign and details around it than if the weather were clear and you could quickly and easily read the sign and then look back at the road.

    You’re example of using a helvetica font for an organic product is still supported by this alternative interpretation. However, this interpretation would go further by implying using helvetica in that situation would probably not significantly increase the time required to recognize the letterforms and words. Therefore it wouldn’t even be an advantage to use an innapropriate font. Also you’re observation that their was a difference between the captive audience in the study and an audience that can simply choose not to spend the attention and time is definitely on point and more significant than any other disconfirming principle.