Perhaps I'm misinterpreting your remarks here, but I would consider
such models (which I agree students should know) as one of the
essential aspects of Human Computer Interaction. I may be a bit
touchy, but I get frustrated when people associate HCI with just
screen design or interface design.
Robin Jeffries
|From: [EMAIL PROTECTED]
|Subject: Re: PPIG discuss: What should SEs know about psych?
|To: Janice Singer <[EMAIL PROTECTED]>
|Cc: [EMAIL PROTECTED]
|MIME-version: 1.0
|Content-transfer-encoding: 7BIT
|Comments: Authenticated sender is <[EMAIL PROTECTED]>
|
|> I'm working with a group of software engineers that is currently
gathering a
|> 'body of knowledge' related to software engineering. The idea is
that this
|> body of knowledge will eventually lead curriculum in software
engineering.
|> They have so far decided that 'cognitive science' would be a useful
topic
|> (although, I myself am not so sure of that), but beyond this, what
can I
|> propose to this group as to what would be useful to take from
psychology.
|Donald Norman's work is significant here.
|I teach my students about different mental models of the system. The
|overlap (congruence) between the task-action mapping model (how the
|user understands what they have to do in order to get the system to
|perform ceratin actions) and the conceptual model (how the user
|perceives the form and purpose of the system) must be high to achieve
|a high degree of uasability in system design.
| ------------- (Venn diagram of
| | T-A |----- congruence between
| | model | | different mental
| -------------- | models of system)
| | Conceptual |
| | model |
| ----------------
|
|THIS IS NOT LIMITED TO INTERFACE DESIGN, but encompasses every
|aspect of how the system works and is configured, in terms of how
humans will
|interact with it. DN's concept of AFFORDANCE (designing so that
|purpose & function is communicated through form) is critical.
|I quote DN's example of Frig/Freezer controls, which are designed as
|separate slider-scales, whereas the two controls are interdependent.
|This is NOT just an interface choice - the choice to make the two
|controls interdependent at the design stage has significant
|implications for how easily users will be able to control the
|artifact, yet this aspect is ignored until the time comes to design
|the interface. At this point the interface is designed misleadingly,
|adding to the confusion, as there is no chance that the user will
|develop congruent task-action mapping and conceptual models of the
|device.
|REFS:
|- See Donald Norman (1988), 'The Psychology of Everyday Things',
Basic
|Books
|- Some of these ideas have come from other authors, such as Harold
|Thimbleby and Paul Booth.
|
|Regards,
| Susan Gasson
|Visiting Assistant Professor, MIS
|School of Management
|State University of New York
|Binghamton NY 13902-6015
|Tel: (607) 777-2337
|http:// http://bingweb.binghamton.edu/~sgasson/
