When people plan, they do so by constructing a simplified mental representation of a problem that is sufficient to solve it—a process that we refer to as value-guided construal.
An ideal, cognitively limited decision-maker should construe a task so as to balance complexity and utility. Preregistered predictions of this model explain people’s awareness, ability to recall problem elements (obstacles in a maze), confidence in recall ability and behaviour in a process-tracing paradigm, even after controlling for the baseline influence of perception and execution as well as ten alternative mechanisms.
These findings support the hypothesis that people make use of a controlled process of value-guided construal, and that it can help to explain the efficiency of human planning.
More generally, our account provides a framework for further investigating the cognitive mechanisms that are involved in construal.
From a broader perspective, our analysis suggests a deep connection between the control of construals and the acquisition of structured representations like objects and their parts that can be cognitively manipulated which can inform the development of intelligent machines. Future investigation into these and other mechanisms that interface with the control of representations will be crucial for developing a comprehensive theory of flexible and efficient intelligence.
All necessary details are free to read in the PDF version of the article.
Surtil : 