Individual differences in time production should indicate differences in the rate of functioning of an internal clock, assuming the existence of such a clock. And sex differences in time production should reflect a difference in the rate of functioning of that clock between men and women. One way of approaching the data is to compute individual regressions of produced duration (P) on target duration (T), after log transformation, and to derive estimates for the intercept and the slope. One could investigate a sex difference by comparing these estimates for men and women; one could also contrast them by looking at mean log(P). Using such indices, we found a sex difference in time production, female participants having a relatively faster internal clock, making shorter time productions, and having a smaller exponent. The question is whether a sex difference in time production would be found using other methods for analyzing the data: (1) the P/T ratio; (2) an absolute discrepancy (|P-T|) score; and (3) an absolute error (|P-T|/T) score. For the P/T ratio, female participants have a lower mean ratio in comparison to the male participants. In contrast, the |P-T| and |P-T|/T indices seem to be seriously compromised by wide individual differences.
Low Rate Scalar Quantization for Gaussian Sources and Absolute Error