HUMAN COGNITIVE PROCESSES IN COMMAND AND CONTROL PLANNING. 4: EXPLORATION OF MEASURES OF PERFORMANCE FOR PLANNING TASK

PDF

Authors
  1. McCann, C.A.
  2. Essens, P.
Corporate Authors
Defence and Civil Inst of Environmental Medicine, Downsview ONT (CAN)
Abstract
The report addresses the continued development of an experimental task for studying cognitive processes in planning. The task is a spatial resource allocation problem in which subjects plan co-ordinated routes for the efficient pickup of items in a store, subject to certain constraints. The report describes changes to a planning task intended to tighten the control of the task through an additional constraint, and the experimental procedure, and to explore potential measures of performance. An experiment was carried out to investigate (a) the difference in planning performance, if any, across problems that had been constructed to be topologically equivalent; and (b) the effect of repeated exposure to these problems on the following measures of performance: time to plan, the quality of the plan, plan similarity, plan faults, number of replans, and planning strategies. Subjects were able to achieve plans within the new constraint (on robot ravel time) in 63% of the trials, and were more successful on some problems than others. Success rate in this regard rose over blocks. In many instances, the requested constraint was violated as a by-product of faults in planning or plan execution. Initially, strategies dealt with the constraints (e.g., robot speed, capacity) individually; more complex strategies adopted in later trials integrated several constraints. TRUNCATED
Report Number
DCIEM-94-61 —
Date of publication
01 Oct 1995
Number of Pages
31
DSTKIM No
95-01480
CANDIS No
150660
Format(s):
Hardcopy;Document Image stored on Optical Disk

Permanent link

Document 1 of 1

Date modified: