Is an accurate verbal language necessary for heart rate discrimination?

Three groups of subjects were required to discriminate levels of cardiac function over three sessions. Two groups were required to detect in which of two contiguous 4-sec periods their heart rate (HR) was slowest. One of these groups was given knowledge of results concerning the correctness of their decisions. Their performance improved significantly. The other group was not given knowledge of results and their performance hovered around chance level. The third group was told to detect the period in which their heart rate was fastest, but they were told 'correct', via knowledge of results, only if they had picked their slowest rate. Their performance was inferior to the knowledge of results group, but superior to the no-knowledge group. They showed, however, little evidence of improved performance over sessions. The results are discussed in terms of a verbal labelling control of required discriminations.