Processing of Numbers and of Physical Magnitude

3 experiments are described in which Ss carried out a decisionmaking task. In all tasks, d' was 0.95 and signal probability was 0.25. In 2 experiments the evidence was presented in the form of numbers, in the third as lengths of rods. Ss performed well with rods, less well with numbers when they did not have knowledge of results, and very poorly with numbers when they did have knowledge of results. It is suggested that many people may not process numbers as a continuum, and may have unclear ideas about probability.