ABSTRACT


Construal Processes in Preference Assessment

Baruch Fischhoff
Shane Frederick
Ned Welch
Carnegie Mellon University

People make choices all the time in their everyday lives, for tasks of
greater and lesser familiarity.  If asked politely, they often will
answer the (often necessarily unfamiliar) tasks created by experimenters
and survey researchers.  Our ability to interpret their responses is
bounded by our ability to interpret how they have construed their tasks.

With rich "real world" tasks, we especially need to figure out which of
the many task features they have chosen to consider and how well those
were understood.  With stylized research tasks, we especially need to
figure out what features they added in order to make them real enough to
answer.  A complete account of these construal processes would consider
both how people choose which lines to read in complex environments and
how they read between the lines in impoverished ones.  

Much of our research progress has come from piecemeal recognition of
ways in which people are more or less sophisticated than we had thought.

The talk will consider the opportunities for studying construal
processes per se, both as a way to increase our theoretical insight and
to assess our current level of attainment.  It will present several
methods for studying these processes, along will some results.  It will
discuss the implications of misunderstanding people's task construals,
when research is conducted for scientific and for practical purposes.