The assessment of discriminative encoding in memory for paired associates

A technique for providing for the measurement of encoding stability and intralist stimulus discrimination is presented and applied to the analysis of some paired-associate data. Essentially, the measure involves a recall test for all stimuli in the list except one, then presenting either a previously tested stimulus or the missing stimulus. Recognition of repeated stimuli reflects code stability, and recognition of the missing stimulus reflects stimulus discrimination.     

The correlation between cue differentiation and associative recall

Two experiments were conducted to correlate performance on a test of cue differentiation with cued recall. In one experiment, subjects were given 2, 5, 8, or 12 study repetitions on verbal items consisting of three-word cues and single-word targets. The cues had varying numbers and patterns of identical elements. Cues were tested for differentiability and for their ability to produce target recall. In the second experiment, study repetitions on the same type of lists were followed by the two tests after the second, fourth, and final study repetitions. With a few exceptions, the frequency of occurrence of items that both were successfully differentiated and produced correct recall did not deviate significantly from predictions based on the assumption of independence of the two measures.     

Transfer of stimulus differentiation and perceptual selection rules

Three experiments examined the conditions under which transfer of stimulus differentiation occurs. Two experiments involved transfer in paired associate lists having identical stimuli and re-paired response terms compared with lists having different stimuli. In one experiment, transfer was determined for lists of three-word stimuli in which various elements were shared among the stimuli. The pattern of redundancy was such that the spatial position of a discriminative element was either consistent or not consistent in different items, or there was no discriminative element. Transfer of differentiation was obtained only when position was constant. In the second experiment, stimuli were single words and were either unrelated, rhymed, or were instances of common concepts. No evidence of transfer of differentiation was obtained with either rhymes or concepts. In the third experiment, subjects learned pairs in which the stimuli were single letters; then subjects transferred to a list in which either rhyming or unrelated stimuli began with the same letters. Transfer was not differential.     

Cue discrimination and forgetting

Subjects learned paired associate fists in which word triads were paired with nouns to a partial criterion; they were then given both an immediate and a 1-week delayed recognition test for discrimination among the triads and a cued recall test. Recall after 1 week was slightly poorer than that occurring immediately, but it was unaffected by various patterns of element identity among the triads. Discrimination improved somewhat over the 1-week interval. Further experiments showed that the immediate test was responsible for both reducing forgetting and improving discrimination, but both effects were independent of identity structure. Forgetting cannot be attributed to loss of differentiation of cues.     

Interference among memory traces

Several sources of interference in memory are identified. These sources may be grossly classified as processing interference, i.e., that due to disruption of whatever activity occurs during the input or output of to-be-remembered material, or trace interaction, i.e., that due to interference among the stored memories themselves. The latter would appear to be due to simultaneous activation of correct and incorrect associations mediated by confusion among cue stimuli. A consideration of the means by which interference is reduced suggests that interfering associates are not weakened, unlearned, or suppressed except possibly when nominal stimuli are identical and sets of target and interfering items are temporally discriminable. Discriminative encoding of cue stimuli may eliminate these associations, if it operates at the perceptual level. Otherwise, potential interfering associates are activated, but may be rendered functionally impotent by discriminating them from correct associations on the basis of either backward association with discriminative stimulus attributes or differential contextual attributes such as frequency, time, order, and strength.     

Stimulus structure, discrimination, and interference.

Subjects learned lists of consonant trigram word pairs varying in intralist formal stimulus similarity. On test trials, subjects were tested with all but one stimulus, then presented either the missing stimulus or one previously tested. Subjects had to correctly identify this stimulus, then recall its associated word. Both recall and recognition performance varied with the difficulty of discovering and utilizing a discriminated element. However, when subjects were instructed as to list structure, recall was improved without improvement in recognition in one condition, and recognition was improved without recall in another. This indicates that stimulus discrimination is neither necessary nor sufficient to reduce intralist interference.