Thomas,

>>I was hoping to do a variant of the Wason selection problem,
>>using C if statements (I ran a trial experiment last year with
>>half a dozen people).
>
>I have to wonder why. Are you looking at positive bias? Are you familiar with the 
>work by Barbee Teasley

I take it you mean:
"Why software testing is sometimes ineffective: two applied studies of
positive test strategy", Journal of applied psychology 1994, Vol 79 pp 142-155.

I think: "The influence of expertise on the four-card selection task in
the domain of software testing", 1992, presented at the 64th annual meeting
of the midwestern psychological association, is more applicable.

One of those cases where studies using students might be
applicable to commercial developers.

> and others on positive bias in program testing (and yes, she used professionals)? 
>I'd be interested to know what you will add to that. Or are you looking at something 
>else?

Results from studies using a Wason selection problem (in one of its
forms) suggest that people don't always reason according to the laws
of mathematical logic.  But there does seem to be a pattern to how
they reason.  Working out theories and explanations for this pattern has
kept various groups of psychologists busy for some years.

The studies have used what might be called non-experts in logic,
ie ordinary people (or as ordinary as undergraduates get to be ;-).
There seems to be a feeling that people trained in logic will deliver
the mathematically correct answer.

I don't see it this way.  Training will alter peoples biases towards giving
a mathematically correct answer more often.  But people will still
give the occasional mathematically incorrect answer.  Are there patterns
to these incorrect answers and if so is it the same pattern as seen in
untrained people?  The 'four-card selection' paper above found that
'advanced' computer science students gave more correct answers than
'novice' computer science students, but they all gave plenty of incorrect
answers.

Whatever image they like to project as all powerful logical thinkers,
programmers frequently (unsubstantiated claim: several per minute when
reading code) make logical reasoning mistakes.  My interest is in finding
out if there is a pattern to this mistakes.

An example of the 'non-mathematical' thinking that programmers use.
For my trial experiment I gave people 100 small very problems to solve.
These were automatically generated and randomised, but of course there
appeared to be a pattern to them.  One of the subjects solved a few
problems and then thought he had spotted a pattern.  He then proceeded to
give answers based on matching the problem to the pattern and appropriately
transforming answers already given, to give new answers (which, if you have
that kind of mind, require less effort).  This created a situation common to
reasoning research; subjects are not solving them problem the experimenter
thinks they are solving.



derek

--
Derek M Jones                                           tel: +44 (0) 1252 520 667
Knowledge Software Ltd                            mailto:[EMAIL PROTECTED]
Applications Standards Conformance Testing   http://www.knosof.co.uk



- Automatic footer for [EMAIL PROTECTED] ----------------------------------
To unsubscribe from this list, mail [EMAIL PROTECTED]  unsubscribe discuss
To join the announcements list, mail [EMAIL PROTECTED] subscribe announce
To receive a help file, mail [EMAIL PROTECTED]         help
This list is archived at http://www.mail-archive.com/discuss%40ppig.org/
If you have any problems or questions, please mail [EMAIL PROTECTED]

Reply via email to