Stimulus Control Exerted by Remember Cues Over Processing in Pigeons’ Short-Term Memory

The aim of the present research was to provide further evidence regarding the role of remember cues on pigeon short-term memory. The first two experiments were conducted to determine whether, using a single cuing procedure, control over delayed matching-to-sample (DMTS) performance could be demonstrated by establishing a no cue condition as an implicit cue to forget. In Experiment 1, subjects were provided with training in a procedure where both forget cue (F) and no cues indicated the omission of the comparison stimuli at the end of the delay interval. In Experiment 2, naive subjects were trained with remember (R) and F trials cued from the outset. In both experiments the cuing effect was evaluated with R, F, and no cue probe trials where the cues were presented in the beginning, middle, or end of a 3 sec delay interval. The results revealed that DMTS accuracy on F cued probe trials was significantly lower than that on R cued or no cue probe trials. Performance on R cued and no cue probe trials was statistically equivalent despite both the explicit training of the no cue condition to function as an implicit cue to forget (Experiment 1), and cuing R and F trials from the outset and presenting the no cue trials only during probe testing (Experiment 2). The main aim of the third experiment was to determine whether the attenuating effect of an R cue over a previously presented F cue was the result of conditioning history. The cuing effect was evaluated in three types of single cued trials (R, F, and no cue), and two types of double cued trials (forget-remember, FR, and forget-novel, FN). The results indicated identical performance in FR and FN cued probe trials, suggesting that the novelty of presenting two cues in the retention interval may play a role in determining DMTS performance in double cued probe trials. Performance on double cued and no cue probe trials did not significantly differ from R or F cued probe trials. The evidence is contrary to the hypothesis that, as a result of previous training, a no cue functions as an implicit remember cue