The applicability of NLP to interaction design.

3 Jan 2008 - 5:11pm
6 years ago
2 replies
164 reads
Darlene Arriola

Hi Alex,
I have to admit that I have almost no understanding of NLP, other
than having read about it briefly on Productivity and Success blogs
in the past few months. Here is a quote found on the web (pasted here
mostly for my benefit): "Neuro-Linguistic Programming (NLP) is an
incredibly powerful discipline that enables people to unblock the
structures of human communication and human excellence. By doing so
people can think, communicate and manage themselves, and others, more
effectively. NLP explores the relationships between how we think
(neuro), how we communicate (linguistic) and our patterns of
behaviour and emotion (programmes)." - from the website

I admit that I don't understand it completely enough to rule it out
of my list of interesting topics to investigate. And it might just
add to my toolset as a designer. This does, however, raise a red

As an Interaction Designer, my first and foremost concern is
understanding the user through various user research methods. I try
to understand as best I can how our users are most likely to use our
products, with the time and resources that I have at my disposal.
This includes (but is not limited to) experience in and knowledge of
the field as well as employing Usability Testing methods. I am not
trying to alter or change people's behavior towards these designs.
That's really up to the individual. Rather, in understanding my
users and how they complete the tasks being presented to them, I can
try to optimize the design to the research. I have been surprised
countless times how differently users can react to what I initially
thought was a very carefully designed interface. IMHO it is much more
efficient to focus on what I have control over (the design) than what
I do not have control over (the behavior of the large number of users
that I design for).

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new


3 Jan 2008 - 7:06pm
Lisa Harper

Alex, I saw your post and was briefly excited. But then I saw that the
NLP you were referring to was not Natural Language Processing and I
became disappointed. I know next to nothing of nuerolinguistic
programming, but am vaguely suspicious that it is a pseudoscience
concocted for some sort of marketing purpose. But I show bias, since
I'm a computational linguist by training and have found an interesting
niche in interaction design for language tools. Not language education
or training tools, rather tools for information extraction, machine
translation, dialogue systems, speech-to-speech translation, etc.

Back to NLP -- Natural Language Processing... if you look toward
psychological theories of communication and discourse, it seems there
is very much to offer interaction design as a discipline. Look
particularly at Herb Clark's 1996 Using Language. Such theories
account for how humans build common ground in conversation but also
account for how people detect and repair error.

In essence, a human-machine interaction is a sort of dialogue. You
could say that interaction designers purposefully, and intuitively,
exploit this understanding of human communication to enable a sort of
loose mixed-initiative dialogue. Researchers working in the area of
computational dialogue (e.g., to enable machines to converse with
humans via text or speech dialogue) have long mused that the same
technology driving speech dialogue could drive interaction with a
non-speech or graphical user interface. Mixed modal interfaces such as
auto systems that allow you to adjust the radio or comfort inside the
car using speech and manipulation are probably the first of these
sorts of "dialogue-driven" interfaces close to making it to the
commercial marketplace (see papers relating to the DICO project, for example). Until now, speech control has
largely been based on simple state machine transactions supported by
VoiceXML or other similar formalisms. But richer formalisms supported
by dialogue theories may actually be relevant to guided or other sorts
of specialized interactions.

In any case, if you find you don't understand neurolinguistic
programming, perhaps the water is muddy for some purpose....

Lisa Harper
The MITRE Corporation

On Jan 3, 2008 8:31 AM, Alexander Livingstone <adl.ixda at> wrote:
> Hi,
> I have been thrust into the role of 'ownership' for several engineering
> packages from a venerable engineering firm.
> They have developed, ahh, organically (I think that's the politest way of
> saying it).
> Suffice it to say they need a thorough looking at. I discovered interaction
> design about three months ago and have been devouring books on the subject
> ever since - I am convinced that it is necessary.
> Coincidentally I also recently read about neuro-linguistic programming (NLP)
> which has a wonderfully airy-fairy name, but seems to apply a lot of clever
> psychological
> methods to looking at and altering people's behaviour and understanding.
> It struck me that some of this would be highly applicable to Interaction
> Design - has anyone tried applying NLP methods - or something similar - in
> there designs, or am I barking up the wrong tree?
> Thanks,
> Alex.
> ________________________________________________________________
> *Come to IxDA Interaction08 | Savannah*
> February 8-10, 2008 in Savannah, GA, USA
> Register today:
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at
> Unsubscribe ................
> List Guidelines ............
> List Help ..................

4 Jan 2008 - 6:09am

Hah, I think I need to remember that the world is running out of space for
three letter acronyms!

Natural Language Processing certainly sounds very interesting and in an
engineering world that's abstracted and represented in algebraic variables,
I can certainly see an instant benefit for what I'm doing in understanding
the material (if *nothing* else). Lisa I've taken your recommendation of
Herb Clark's book and will grab it from Amazon forthwith.

I suppose my take on Neuro Linguistic Programming (given the limited time
I've spent looking at it) is that its techniques are based upon an model of
the patterns that our minds follow - how we learn, retain information,
behave, etc. which, assuming it is accurate, would be good for our overall
understanding of humans. So in this respect, it might be able to help to
remove a part of the uncertainty
in any design that I undertake. My 'pretend its magic' (can you tell
I've been reading?) aim for what I'm going to be
doing is that I can use behavioural patterns so that out applications can be
to our field
what Derren Brown is to 'mind reading' (If you haven't heard of him, I
suggest you go to Amazon and grab a DVD / unbox video
- it is absolutely fascinating - and great entertainment to
boot!). I gather his methods are partially
rooted in the same sort of theory that Neuro Linguistic Programming is.

It could be that it is a red herring - and human psychology as a whole is
something I just need to look at more. After all, I have virtually no
understanding of any sort of social science - I trained in Physics and Naval
Architecture, so I've a very steep learning curve ahead of me in IxD.

Re-reading this, perhaps I need to temper my enthusiasm and start slowly ( =]

Thanks, all, for the input!


Syndicate content Get the feed