Xref: utzoo sci.psychology:2045 comp.cog-eng:1166
Path: utzoo!utgpu!jarvis.csri.toronto.edu!me!radio.astro!helios!utpsych!raymond
From: raymond@utpsych.toronto.edu (Raymond Shaw)
Newsgroups: sci.psychology,comp.cog-eng
Subject: Re: navigating through menus with colour
Summary: just adding a little psychology to the discussion
Keywords: colour, menus
Message-ID: <1989Jun6.184722.13613@utpsych.toronto.edu>
Date: 6 Jun 89 18:47:22 GMT
References: <572@hfserver.hfnet.bt.co.uk> <10116@watcgl.waterloo.edu>
Reply-To: raymond@psych.toronto.edu (Raymond Shaw)
Organization: Psychology, U. of Toronto
Lines: 42

In article <10116@watcgl.waterloo.edu> mejanzen@watcgl.waterloo.edu (Michael E. Janzen) writes:
>In article <572@hfserver.hfnet.bt.co.uk>, davet@hfserver.hfnet.bt.co.uk (David Travis) writes:

Travis:
>> One problem with menu-driven interfaces is that they encourage users to
>> be lazy: if you ask them to recall the options on a pull-down menu,
>> performance is very poor. Options are recalled by their approximate
   ^^^^^^^^^^^                           ^^^^^^^^
Just out of curiosity, just what is your measure of performance?  When you
say "recall" do you mean that you ask your subjects to tell you the names
of the options, or to describe them?  If you ask them to name them, how
stringent are you at scoring their answers?  In my own experience with menus,
after enough practice, I find that I know both the names and positions of the
options, so that I can often get ahead of the software in keystrokes.  So
practice is an important issue as well.

>> spatial position in the menu. A prediction from this would be that
>> colour-coding menu items would improve search times. Has there been any
>> work produced on navigating through menus using colour as a cue? 

This suggests that you aren't asking them to "recall" the items, but instead
are asking them to find items in the menu from some starting position.  In
that case, you probably have too many items on your menus for easy discrimina-
tion.

Janzen:
>I am taking part in research on that topic.  It appears that colour is more
>effective than position in improving speed and accuracy of character 
>recognition, but combing the two is particularly effective.
 ^^^^^^^^^^^
Is it recognition, or discrimination?  Is your measure choice? or naming?

It seems to me that using a menu taps some sort of procedural memory, whereas
naming or recalling the options in a menu taps some sort of episodic/semantic
memory.  Generally, researchers seem to find dissociations between these
"memory systems," so it seems likely that the nature of your measure will
affect the nature of the performance you are likely to obtain, particularly
if the way you train your subjects to use the menus is different from the
way you test them.

-Ray Shaw
raymond@psych.toronto.edu