A Generalized Model of Temporal Motor Control Subject to Movement Constraints

Abstract

First- and second-ordar linear models relating mean arm movement time to Fitts'a index of difficulty (ID) variable and a lateral movement constraint variable were formulated and validated experimentally for a number of alternative types of constraints using 10 undergraduate male students as subjects. The first-order models were found to explain from 67 % to 84 % of the variation in mean movement time while the corresponding figures for the second-order models were 90% and 99·2%. The information rates generated by the tasks ranged from 0·5 to 3·0 bits s−1 between the different types of lateral constraints, which were from 50% to 94% of those generated by manipulative hand movements (with fixed arm) subject to the same types of movement constraints. The substantial performance variations that occurred among individual subjects were significantly (at the 2% level) and positively influenced by task difficulty as measured by the two ID variables. The experimental movement times showed pronounced deviations from predetermined motion-time systems predictions.