Error of Mental Models? LO1959

kent_myers@smtplink.sra.com
Thu, 06 Jul 95 10:48:09 EST

Reply to McMaster, LO1947

Your objections to mental modeling are to its abuse, primarily to the
creation of neat formulas. But some portion of a person's thinking IS
formulaic. If the formula is creating errors, then that's worth knowing.

Using the mental model approach, a human factors researcher described a
discrepancy between how railroad engineers think a locomotive operates and
the demonstrable way in which the locomotive operates. The difference in
models helped explain a series of deadly errors that had been dismissed as
"human error" and which led to two favorite remedies: prosecute the
perpetrators, and train more in the same way. Under the modeling
approach, training was changed to add models midway between general
physics and control behaviors. In unusual and emergency situations
habituated control behaviors are put aside and the operator acts on the
basis of how he thinks the machinery works (describable as a mental
model), often bypassing the control behaviors listed in the manual.

I don't see what's wrong with this restricted use. I don't need to assert
that a mental model "exists", only that I can use a model to usefully
describe how a person thinks and acts in a restricted domain.

The domain can be quite broad, depending on how broadly a person uses
formulas. Cognitive therapy describes "thinking errors" and their
correction. That therapy, unlike most, takes a person's explanation of
his actions seriously. It also helps people, and does so in a way that
the person understands. If a formula is applied in a moral domain, the
therapist doesn't have to say what is right when pointing out a thinking
gap, nor does the therapist have to replace one formula with another.

--
Kent Myers          kent_myers@smtplink.sra.com