Stimulus encoding selectivity at high degrees of paired-associate learning

Sixty-two Ss learned six CCC trigram-digit pairs by the study-test method; 21 Ss continued the PA acquisition trials to a learning criterion of one errorless trial, while 41 Ss were carded to 300% overlearning or 24 trials, whichever came first. The Ss were then shown individual letters of the trigrams, with position cues provided, and were tested for their ability to provide the digit and produce the additional letters. Digit recall was much better to initial letters than to those in the second or third positions of the trigrams; with oveflearning, digit recall improved to the initial letters but not to letters in second and third positions. Production of additional letters of the trigrams was not appreciably related to either cue position or degree of PA learning. There was some evidence ofincreased selectivity with higher degrees of PA learning.     

Part-list recall following part-whole learning

Following five trials of part-list free recall learning and six trials of whole-list free recall learning, Ss were asked to recall the part-list items on two successive trials. Experimental Ss whose whole list contained all of the part-list items recalled more part-list items than did control Ss whose whole list contained none of the part-list items. The data are inconsistent with current notion of subjective organization.     

Arousal and paired-associate learning

Walker and Tarte (1963) postulate that at short retention intervals high arousal paired-associates are reproduced more poorly than low arousal items. Walker and colleagues believe that this hypothesis is confirmed by their paired-associate learning studies. However, results of these paired-associate learning studies are position confounded artifacts. Better recall of low arousal items at short-term retention is caused by the coincidence of the recency effect and low arousal at the end of the trial. When these position effects are controlled there is no action decrement for the high arousal paired-associates. To test this assumption, the Kleinsmith and Kaplan study (1963) was replicated and two other variations were conducted. In these three studies with 76 subjects, which were tested at two minutes or 1 week, the action decrement occurs only when the two position effects coincide.     

Stimulus redundancy and immediate recall

This study examined immediate recall in two stimulus prefix and two stimulus suffix conditions and in a condition that combined a prefix and suffix. Suffixes and the combination of a prefix with a suffix interfered more with recall overall than did prefixes. Performance in each of the conditions that included a prefix was significantly better overall than in appropriate control conditions, in which interference was augmented by a redundant element in recall. It was suggested that prefixes and suffixes lie operationally on a continuum and that their effects result from the subject's inability to dissociate the redundant element from the memory series. However, the location of redundancy imposes different processing requirements that differentially influence recall.     

Second responses in paired-associate learning

Mathematical models of memory can be evaluated using a continuous paired-associate learning paradigm in which a second response is requested immediately following an error. This procedure provides discriminability among the models on the basis of the relationship between the first- and second-response error probabilities. Theoretical predictions are derived for a variety of models, including finite-state and noncontingent operator models, formulations based on the theory of signal detectability, the multicomponent memory model of Bower (1967b), and the long- and short-term model of Atkinson and Shiffrin (1968). None of these is unequivocally acceptable, and many are decisively rejected by the data of a single-trial experiment. The most satisfactory models are the multicomponent model and the Atkinson and Shiffrin model, both of which assume a greater degree of internal structure than the rejected models. It appears that internal structure of this sort will be a necessary part of any satisfactory model of second responses.