The adjusting-delay task introduced by Mazur (Quantitative analyses of behavior: V. The effect of delay and of intervening events on reinforcement value, 1987, pp. 55-73) has been widely used to study choice of delayed reinforcers. This paradigm involves repeated choice between one reinforcer delivered after a fixed delay and another, typically larger, reinforcer delivered after a variable delay; the variable delay is adjusted depending on the subject's choice until an equilibrium point is reached at which the subject is indifferent between the two alternatives. Rats were trained on a version of this task and their behaviour was examined to determine the nature of their sensitivity to the adjusting delay; these analyses included the use of a cross-correlational technique. No clear evidence of sensitivity to the adjusting delay was found. A number of decision rules, some sensitive to the adjusting delay and some not, were simulated and it was observed that some effects usually supposed to be a consequence of delay sensitivity could be generated by delay-independent processes, such as a consistent, unchanging relative preference between the alternatives. Consequently, the use of explicit analysis of delay sensitivity is advocated in future research on delayed reinforcement.