The nictitating membrane response (NMR) of New Zealand white rabbits was conditioned to a tone and infraorbital shock. 32 rabbits served as Ss in Exp. I, and 16 in Exp. II. Experimental Ss received paired CS-UCS presentations together with interpolated UCS alone (UCSa) trials. Controls received the same pattern of paired trials, but no unpaired UCS presentations. For both groups, UCR amplitude on paired trials remained at a high level, but declined progressively on UCSa trials in the experimental group. A parallel human eyelid conditioning experiment using 94 male and female undergraduates showed the opposite result. It is therefore unlikely that failure to observe conditioned diminution of the UCR with the rabbit NMR was due to the use of a within-S design. The use of UCSa trials interfered with conditioning of the NMR, and a control experiment suggested that this interference was not due to massed UCS presentations. (PsycINFO Database Record (c) 2006 APA, all rights reserved)