Mental Models vs/with Personas

25 Jun 2007 - 11:54am
7 years ago
12 replies
950 reads
tdellaringa
2006

I'm struggling a bit with the practical advantage of mental models vs.
personas. I've been doing some reading on MM (and there doesn't seem to be
many good articles, found a handful) and there is quite a difference of
opinion on their effectiveness, their creation and their benefit within an
organization.

What strikes me as that in the end, both MM and personas seem to offer very
similar outputs. They both offer a kind of segmentation and an understanding
of an archetype user which is used as a design tool and so forth.

While I assume there is overlap in user interview data that can be used to
create both tools, it seems that MM are very time consuming to properly
create, just as personas are. What I wonder is there really that much
benefit in having both - particularly in having a MM? Having a persona seems
to offer much of what a MM brings - it allows you to think as the user and
how they see the system. Is having a "cityscape" diagram going to really
help that much more?

My guess is that the more complex the project, the more bang you might get
out of them. When time can be such a limited commodity, I'm wondering if
it's best to go with personas only, if there was a choice to be made (and
possibly even if there isn't).

It's quite possible, however, that I am missing the boat entirely. Too bad
Rosenfeld's book is not out yet.

Love to hear some opinions.

Tom

Comments

25 Jun 2007 - 12:07pm
Lori Landesman
2007

Tom:

Your post interests me because you're describing mental models as something that you would create (I think, if I'm understanding your post correctly). I think of mental models as something users already have -- baggage they bring with them to your design. The more you understand users mental models (or, the way they believe things work as opposed to how those things actually do work), the better able you are to let your design capitalize on them rather than work against them.

So, you can create personas based on your understanding of your users and try to use that to anticipate their mental models and design for them, but I often find that I don't completely understand the mental models my users are bringing to the table until I conduct usability testing on an actual design and hear users talk through them. That's when it becomes clear. Then, I can refine a design with the mental models in mind.

I find then, that you can't really leave out consideration of mental models if you want to understand what users believe is going on in an interface, but that consideration often happens at a different time in the process and in a different context than persona creation.

--Lori

Tom Dell'Aringa <pixelmech at gmail.com> wrote: I'm struggling a bit with the practical advantage of mental models vs.
personas. I've been doing some reading on MM (and there doesn't seem to be
many good articles, found a handful) and there is quite a difference of
opinion on their effectiveness, their creation and their benefit within an
organization.

What strikes me as that in the end, both MM and personas seem to offer very
similar outputs. They both offer a kind of segmentation and an understanding
of an archetype user which is used as a design tool and so forth.

While I assume there is overlap in user interview data that can be used to
create both tools, it seems that MM are very time consuming to properly
create, just as personas are. What I wonder is there really that much
benefit in having both - particularly in having a MM? Having a persona seems
to offer much of what a MM brings - it allows you to think as the user and
how they see the system. Is having a "cityscape" diagram going to really
help that much more?

My guess is that the more complex the project, the more bang you might get
out of them. When time can be such a limited commodity, I'm wondering if
it's best to go with personas only, if there was a choice to be made (and
possibly even if there isn't).

It's quite possible, however, that I am missing the boat entirely. Too bad
Rosenfeld's book is not out yet.

Love to hear some opinions.

Tom
________________________________________________________________
Welcome to the Interaction Design Association (IxDA)!
To post to this list ....... discuss at ixda.org
List Guidelines ............ http://beta.ixda.org/guidelines
List Help .................. http://beta.ixda.org/help
Unsubscribe ................ http://beta.ixda.org/unsubscribe
Questions .................. list at ixda.org
Home ....................... http://beta.ixda.org

---------------------------------
Finding fabulous fares is fun.
Let Yahoo! FareChase search your favorite travel sites to find flight and hotel bargains.

25 Jun 2007 - 1:02pm
tdellaringa
2006

On 6/25/07, Lori Landesman <loriland910 at yahoo.com> wrote:
>
> Your post interests me because you're describing mental models as
> something that you would create (I think, if I'm understanding your post
> correctly). I think of mental models as something users already have --
> baggage they bring with them to your design. The more you understand users
> mental models (or, the way they believe things work as opposed to how those
> things actually do work), the better able you are to let your design
> capitalize on them rather than work against them.

Right, I probably worded my question wrong. What I am getting at is the
research behind, and the deliverable for mental models. So, how different
(if at all) are user interviews for mental models vs. personas? Can the same
data be used? And lastly, is the output, the chart or whatever (something
like this<http://www.rosenfeldmedia.com/books/mental-models/blog/moviegoer_alignment_diagram/index.php>)
basically being used for many of the same things personas are? If not, how
are they different in a practical sense? To research and quantify mental
models is a lot of work, and I guess I'm looking for the payoff.

So, you can create personas based on your understanding of your users and
> try to use that to anticipate their mental models and design for them, but I
> often find that I don't completely understand the mental models my users are
> bringing to the table until I conduct usability testing on an actual design
> and hear users talk through them. That's when it becomes clear. Then, I can
> refine a design with the mental models in mind.

See, here you are talking about mental models, but how are you manifesting
them, other than forming them in your own mind? If you only form them in
your own mind, how do you keep it separate from your own model, or desired
models (or goals)? If you are only "keeping them in mind" how does it
benefit the rest of the team? (Maybe it doesn't?)

I find then, that you can't really leave out consideration of mental models
> if you want to understand what users believe is going on in an interface,
> but that consideration often happens at a different time in the process and
> in a different context than persona creation.
>

See, I think I am getting a picture of the user's mental model throughout
the user interview and persona building process. I am certainly considering
how they think specifically how a system or site works, and acting
accordingly. I think some of this is intimated in personas and scenarios.

Rosenfeld talks about a 6 week timeline to "make" mental models. (In fact,
throughout his site the verbage "create" and "make" is used in regard to
MMs). That is a lot of money and time to spend for something that seems to
be in the UX community something of a black art. If it is somewhat of a
black art among ourselves, that makes it that much harder to legitimize it
to my internal clients who don't want to spend 6 weeks for something of
which they cannot see the benefit.

Maybe it isn't a black art, but I am having a hard time finding anything
really solid on MMs. Most of it is theory or discussion. And so far I've
really only found one deliverable, Rosenfelds. And it clearly took a lot of
time to produce.

Tom

26 Jun 2007 - 4:03pm
natekendrick
2005

I believe mental models are the result of user interviews > personas
> scenarios.

the greater process is this: user interviews > personas > scenarios >
user flows > mental models > task model > information architecture

many times we skip the mental model diagram simply because the client
doesn't engage at that level. They engage at a site IA diagram level.

But it is important for the design team - the act of trying to
distill mental models into an illustrative, understandable diagram
helps crystallize what is in your head and move onto the IA

And its all about communicating your design in the end.

On Jun 25, 2007, at 11:02 AM, Tom Dell'Aringa wrote:

> On 6/25/07, Lori Landesman <loriland910 at yahoo.com> wrote:
>>
>> Your post interests me because you're describing mental models as
>> something that you would create (I think, if I'm understanding
>> your post
>> correctly). I think of mental models as something users already
>> have --
>> baggage they bring with them to your design. The more you
>> understand users
>> mental models (or, the way they believe things work as opposed to
>> how those
>> things actually do work), the better able you are to let your design
>> capitalize on them rather than work against them.
>
>
> Right, I probably worded my question wrong. What I am getting at is
> the
> research behind, and the deliverable for mental models. So, how
> different
> (if at all) are user interviews for mental models vs. personas? Can
> the same
> data be used? And lastly, is the output, the chart or whatever
> (something
> like this<http://www.rosenfeldmedia.com/books/mental-models/blog/
> moviegoer_alignment_diagram/index.php>)
> basically being used for many of the same things personas are? If
> not, how
> are they different in a practical sense? To research and quantify
> mental
> models is a lot of work, and I guess I'm looking for the payoff.
>
> So, you can create personas based on your understanding of your
> users and
>> try to use that to anticipate their mental models and design for
>> them, but I
>> often find that I don't completely understand the mental models my
>> users are
>> bringing to the table until I conduct usability testing on an
>> actual design
>> and hear users talk through them. That's when it becomes clear.
>> Then, I can
>> refine a design with the mental models in mind.
>
>
> See, here you are talking about mental models, but how are you
> manifesting
> them, other than forming them in your own mind? If you only form
> them in
> your own mind, how do you keep it separate from your own model, or
> desired
> models (or goals)? If you are only "keeping them in mind" how does it
> benefit the rest of the team? (Maybe it doesn't?)
>
> I find then, that you can't really leave out consideration of
> mental models
>> if you want to understand what users believe is going on in an
>> interface,
>> but that consideration often happens at a different time in the
>> process and
>> in a different context than persona creation.
>>
>
> See, I think I am getting a picture of the user's mental model
> throughout
> the user interview and persona building process. I am certainly
> considering
> how they think specifically how a system or site works, and acting
> accordingly. I think some of this is intimated in personas and
> scenarios.
>
> Rosenfeld talks about a 6 week timeline to "make" mental models.
> (In fact,
> throughout his site the verbage "create" and "make" is used in
> regard to
> MMs). That is a lot of money and time to spend for something that
> seems to
> be in the UX community something of a black art. If it is somewhat
> of a
> black art among ourselves, that makes it that much harder to
> legitimize it
> to my internal clients who don't want to spend 6 weeks for
> something of
> which they cannot see the benefit.
>
> Maybe it isn't a black art, but I am having a hard time finding
> anything
> really solid on MMs. Most of it is theory or discussion. And so far
> I've
> really only found one deliverable, Rosenfelds. And it clearly took
> a lot of
> time to produce.
>
> Tom
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> List Guidelines ............ http://beta.ixda.org/guidelines
> List Help .................. http://beta.ixda.org/help
> Unsubscribe ................ http://beta.ixda.org/unsubscribe
> Questions .................. list at ixda.org
> Home ....................... http://beta.ixda.org

26 Jun 2007 - 4:13pm
tdellaringa
2006

On 6/26/07, Nathan Kendrick <natekendrick at yahoo.com> wrote:
>
> But it is important for the design team - the act of trying to
> distill mental models into an illustrative, understandable diagram
> helps crystallize what is in your head and move onto the IA

Yep, that's the heart of the matter I am trying to get to. Yet the only task
model I can find - which seems to be the output of MM research, is the one I
posted. There has to be other ways to do it, but I can't find them.

Tom

26 Jun 2007 - 4:53pm
natekendrick
2005

I don't think its an either/or between an MM diagram and Personas.

I believe developing Personas is more required than a MM diagram. In
my experience the mental model diagram helps get the client out of
their own business perspective and not much more. The value I get is
much more, and there is always time for it at some level (like in a
sketchbook).

If you mean Personas vs. a Task Model - then yes, it could be an
either/or situation if time/budget only allows for one.

imo - a task model is ultimately more useful. Balance that out with
the fact your client is probably requesting Personas and you have to
consider the fact you'll need to educate your client on the vagaries
of use of these discrete design deliverables. In the end its
probably not worth the debate... do the one that makes sense given
the project.

Sometimes the task model will be so difficult to get complete and
accurate that Personas and scenarios are sufficient.

On Jun 26, 2007, at 2:13 PM, Tom Dell'Aringa wrote:

> Yep, that's the heart of the matter I am trying to get to. Yet the
> only task model I can find - which seems to be the output of MM
> research, is the one I posted. There has to be other ways to do it,
> but I can't find them.

26 Jun 2007 - 7:32pm
Dan Saffer
2003

On Jun 26, 2007, at 2:53 PM, Nathan Kendrick wrote:

>
> I don't think its an either/or between an MM diagram and Personas.

It's not. They are separate tools for different purposes. Personas
are for archetypal users. MM diagrams are for mapping an activity,
with the tasks and decisions that need to be made, aligned with the
objects, features, and content that suppose those tasks and
decisions. I prefer the term "Alignment Diagram" myself, to avoid the
nomenclature confusion we had earlier in this thread.

Todd Warfel's Task Analysis Grid

<http://toddwarfel.com/?p=16>

is an interesting mix of the two things.

Dan

27 Jun 2007 - 8:47am
tdellaringa
2006

On 6/26/07, Dan Saffer <dan at odannyboy.com> wrote:

> On Jun 26, 2007, at 2:53 PM, Nathan Kendrick wrote:
> > I don't think its an either/or between an MM diagram and Personas.
>
> It's not. They are separate tools for different purposes. Personas
> are for archetypal users. MM diagrams are for mapping an activity,
> with the tasks and decisions that need to be made, aligned with the
> objects, features, and content that suppose those tasks and
> decisions. I prefer the term "Alignment Diagram" myself, to avoid the
> nomenclature confusion we had earlier in this thread.
>
> Todd Warfel's Task Analysis Grid
>
> <http://toddwarfel.com/?p=16>
>
> is an interesting mix of the two things.

So I'm clear, you are saying it's a mix of Personas and MMs? I've seen this
grid before, he explains it as a visual representation of project
requirements. It seems based on the excel file, which is more of a task
analysis. I'm not quite getting were MMs fits into this grid.

Tom

27 Jun 2007 - 9:16am
Amy Silvers
2007

Alignment diagrams (I prefer that term, and it's what Indi Young,
who's written extensively on the subject, is using nowadays) go a
step farther than task analysis by mapping the tasks you've
identified them to the content and functionality/tools that support
them.

Among other things, it's a very useful way to do gap analysis when
you're doing a large redesign. It's also a great way to give
stakeholders (internal and external) a big-picture view of how
existing content and features do and don't support users' goals.

The enormous size of the diagrams may seem unwieldy, but it serves a
purpose. We've done sessions with stakeholders where we tape the
diagram up around the room and have them scrutinize it, add to it,
brainstorm from it, etc. We use personas throughout the design
process, and we don't use alignment diagrams on every project, but
they're a great tool when appropriate.

--Amy

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://beta.ixda.org/discuss?post=17584

27 Jun 2007 - 9:45am
Dan Saffer
2003

On Jun 27, 2007, at 6:47 AM, Tom Dell'Aringa wrote:

>
> So I'm clear, you are saying it's a mix of Personas and MMs?

Todd's document is a step towards a mix of persona, scenarios,
requirements document, and MM/Task Analysis. It tries to align all
those things (and does a pretty good job of it). A traditional task
analysis is just a raw list of task, not shown in any context. These
documents put the context back in.

Dan

27 Jun 2007 - 9:47am
tdellaringa
2006

That makes a little more sense to me Amy, and I can see how that can
be really valuable.

I think the problem I am having is that I'm looking for a concrete
item for mental models, and maybe there isn't one - almost like
looking for a concrete item for user interviews.

We use the data from user interviews to form various deliverables,
like personas, usability goals and functional requirements. But there
is no "thing" that is user interviews (discounting the notes).

I guess it's the same for MMs. You get a picture of what that is
based on user research, and that information is used in task analysis
and further in these affinity diagrams. But again, there is no
"thing" that is a MM, it's simply information gained from users
that helps drive design.

Hope I'm getting that right.

Tom

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://beta.ixda.org/discuss?post=17584

27 Jun 2007 - 2:26pm
Joannes Vandermeulen
2006

Tom

In my design practice, a %u2018target mental model%u2019 is the
keystone deliverable that finishes up the analysis and initiates the
design. It%u2019s the deliverable with the highest level of
abstraction. In this 2004 poster
(www.namahn.com/FileZone/Namahn-poster.pdf ) we call it
%u2018conceptual model%u2019 but a more appropriate name would be
%u2018target mental model%u2019, thus avoiding confusion with the
conceptual design, which comes next.

For formalisms and tools, we%u2019ve tried a lot, but mindmaps are
well received. Entity-Relationship diagrams are a pretty good fit as
well.

I see the relationship between the technical model, the target mental
model (tMM) and the instantiated/individual/idiosyncratic mental model
(iMM) as follows, also expressed in this diagram
(www.namahn.com/FileZone/mentalModelsDiagram.gif ):

- The product can be understood from a technical viewpoint (down to
the atom level, if you wish to go that deep), but users mostly
don%u2019t need, don't want or cannot handle that kind of
understanding. That%u2019s the technical model.
- It's a designer task to create a target mental model that reveals
the useful functions of the product, while also a being a good match
to the %u2018baggage%u2019 (indeed, Lori) the user already carries.
Metaphor and analogy work well here. Also, different user groups get
different, but overlapping target mental models.
- Next, you lose control, when the wildly interpreting user makes
what she want from your target mental model. You can only hope that
the individual mental models resemble the target mental model. If
not, the help desk is in trouble.

Here%u2019s a picture of a target mental model that I did with
students in Germany a while ago. Ignore the content (some of it in
German) but consider the formalism. It%u2019s quite simple and
powerful (www.namahn.com/FileZone/20070612-161437.jpg )

I agree that the moviegoer 'alignment diagram' is best called
exactly that, and not a mental model.

There may be synergy between the creation of personas and mental
models, but they are distinct.

hope this helps

Joannes Vandermeulen
jv at namahn.com

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://beta.ixda.org/discuss?post=17584

27 Jun 2007 - 2:42pm
Joannes Vandermeulen
2006

--- sorry about the quotes coming up as computer codes in my previous
post. I'd rather repost. --

In my design practice, a 'target mental model' is the keystone
deliverable that finishes up the analysis and initiates the design.
It's the deliverable with the highest level of abstraction. In this
2004 poster (www.namahn.com/FileZone/Namahn-poster.pdf ) we call it
'conceptual model' but a more appropriate name would be 'target
mental model', thus avoiding confusion with the conceptual design,
which comes next.

For formalisms and tools, we've tried a lot, but mindmaps are well
received. Entity-Relationship diagrams are a pretty good fit as well.

I see the relationship between the technical model, the target mental
model (tMM) and the instantiated/individual/idiosyncratic mental model
(iMM) as follows, also expressed in this diagram
(www.namahn.com/FileZone/mentalModelsDiagram.gif ):

- The product can be understood from a technical viewpoint (down to
the atom level, if you wish to go that deep), but users mostly don't
need, don't want or cannot handle that kind of understanding. That's
the technical model.
- It's a designer task to create a target mental model that reveals
the useful functions of the product, while also a being a good match
to the 'baggage' (indeed, Lori) the user already carries. Metaphor
and analogy work well here. Also, different user groups get
different, but overlapping target mental models.
- Next, you lose control, when the wildly interpreting user makes
what she want from your target mental model. You can only hope that
the individual mental models resemble the target mental model. If
not, the help desk is in trouble.

Here's a picture of a target mental model that I did with students
in Germany a while ago. Ignore the content (some of it in German) but
consider the formalism. It's quite simple and powerful
(www.namahn.com/FileZone/20070612-161437.jpg )

I agree that the moviegoer 'alignment diagram' is best called
exactly that, and not a mental model.

There may be synergy between the creation of personas and mental
models, but they are distinct.

hope this helps

Joannes Vandermeulen
jv at namahn.com

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://beta.ixda.org/discuss?post=17584

Syndicate content Get the feed