ABSTRACT
Subjects (N = 25) competed for one monetary prize in a group session by estimating time intervals, ranging from 1.7 to 12.1 s. They were told that the one with the smallest sum error of estimates would be rewarded with a monetary prize. They marked each estimate on a response sheet and, subsequently, read and marked the duration of each interval (the correct response) from a stopwatch. Reading errors were more often in the direction of the respective estimate error than in the opposite direction and, hence, in the direction of the subject's advantage. This was particularly so for difficult trials, that is, when the watch hand had stopped between tenths of a second.