This study investigated the effects of between-trials variability on the alternation discrimination of retardates. Brightness and size were variable between trials and irrelevant. Initial response outcome and mental age were also independent variables. Neither between-trials variability nor mental age produced a significant main effect, while the main effect of initial response outcome was significant. Non reinforcement of the first response was associated with significantly more nonalternation responses than was the condition of reinforcement for the first response. The results were discussed in terms of the Zeaman-House Attention theory and subject response predispositions. (Author)