Classes and collections: Internal organization and resulting holistic properties

This work presents an analysis of a type of concept, the collection, not readily characterized by class inclusion models. Collections, the referents of collective nouns (e.g., pile, family, bunch), are argued to differ from classes in (a) how membership can be determined, (b) part-whole relationships, (c) internal structure, and (d) the nature of the higher order units they form. From this analysis, it is hypothesized that the psychological integrity of collections is greater than that of classes. Collections and objects, in contrast to classes, both require specified relationships among the parts and both result in a coherent psychological unit. It was suggested that objects form a relatively more stable unit than collections. Corresponding to this analysis the degree of psychological integrity is hypothesized to lead to different degrees of difficulty in making part-whole comparisons for objects, collections, and classes in modified Piagetian class-inclusion paradigms. The hypothesized difference in performance was found for collections and classes and an alternative linguistic explanation for the greater success with collections was eliminated. However, children performed equally well on tasks involving collections and objects, raising the possibility that when elements are organized into collections, they form psychological units which are as coherent as objects.     

Retrieval processes in recall

A reaction time (RT) paradigm was developed to study retrieval processes in paired associate (PA) recall. Prior to the experimental session, subjects learned lists of PAs (varying in length from three to 24 pairs); during the experimental session, subjects' RT to say the response word from the onset of a visually presented stimulus word was measured. The implications of several classes of retrieval models were discussed in the context of this paradigm. The shape of the RT-list length function, practice, and sequential effects were all of interest in distinguishing among models. Four experiments were reported which were designed to (1) establish the baseline effects in this paradigm, (2) determine which of these effects should be attributed to the retrieval stage of processing, and (3) investigate the effect of semantic memory in this task. Results suggest that simple serial scanning models are inadequate to handle the data from this task. Strength, direct-access, or parallel processing models, on the other hand, seem to capture the qualitative effects present in our experiments. When a strength model was formalized and fit to the data from Experiment I, it was found that a two-trace version gave good quantitative fits while a one-trace version did not, suggesting that both short- and long-term memory independently contribute in this task.     

The Premack principle in human experimental and applied settings

Studies claiming support for the Premack Principle in human experimental and applied settings are reviewed in terms of the standard test conditions associated with the Principle and in terms of the Principle's conceptual constrains. Little definitive evidence was found to support the claim that a high probability response will reinforce a lower probability response, nor was the converse contention supported, that a low probability response will act as a punisher for a high probability response. More importantly, among the reports reviewed there was no evidence for the reversibility of the reinforcement relationship.     

Covariation and temporal contiguity as principles of causal inference in young children

Four- to seven-year-old children observed a simple physical effect which could be attributed to either a consistent but noncontiguous covariate or a contiguous but inconsistent covariate. When there was a physical rationale for the temporal delay between covariate and effect, children attributed the effect to the consistent but noncontiguous covariate. In the absence of such a rationale, they attributed the effect to the contiguous but inconsistent covariate. It was concluded that neither consistent covariation nor strict temporal contiguity were essential aspects of causal inference for these children.     

Children's understanding of sentences with contingent relations: Why are temporal and conditional connectives so difficult?

Children aged 5-, 7-, and 9 were tested on two different tasks to assess their understanding of sentences containing the connectives “when,” “as soon as,” “before,” “after,” “if,” “if-not,” “unless,” and “unless-not.” The results revealed significant improvement between the ages of 5 and 7. This improvement was attributed, in part, to an increased ability to retain information expressed in subordinate clauses. At all age levels, however, the connectives “unless” and “unless-not” were consistently misinterpreted; “unless” was treated as its polar opposite “if,” and “unless-not” was treated as its polar opposite “if-not.”     

A note on the use of A′ as a measure of sensitivity

The use of A′, a nonparametric measure of sensitivity in tests of recognition memory, is discussed. In particular, two formulas for computing A′ are considered. One formula, which was reported in a recent paper by Berch (1975), should be used only when the observed hit rate is greater than or equal to the false alarm rate. A second formula is presented for computing A′ when the hit rate is less than the false alarm rate.     

Social influence via threats

Influence via threats was investigated in correspondent relationships (characterized by commonality of interest) and in noncorrespondent relationships (characterized by conflict of interest). In addition, the degree of informational power attributed by the influencee to the influencer and to himself, and the severity of threats used by the influencer were manipulated. The results point to the existence of two qualitatively different processes of influence. In correspondent relationships, threats are viewed as conveying the influencer's belief about the choice of action that would benefit both parties to the relationship. Further, the more severe the threat, the more confident the influencer would appear to be of his choice of action. As a result, the influencee's willingness to rely on the influencer's judgment, as conveyed by the threat, becomes an important aspect of the influence process. Such willingness is affected by the influencee's attribution of informational power to the influencer and to himself. In noncorrespondent relationships, on the other hand, suspicion and mutual distrust preclude reliance on threats as sources of information. Under such circumstances, social influence via threats becomes a simple process of coercion.     

Misconceptions about the Fishbein model: Reflections on a study by Songer-Nocks

Songer-Nocks has questioned the generality and sufficiency of Fishbein's model of intentions. It is shown that her conclusions reflect a basic misunderstanding of the model and its implications. Several misconceptions are identified and an attempt is made to clarify some of the theoretical and methodological issues involved.     

Reply to Fishbein and Ajzen

The purpose of the Songer-Nocks study was clarified, and the relationship of the predictions and results to the theoretical formulations of Fishbein and Ajzen were discussed. Results from the model-testing procedure originally performed with behavior as the criterion were presented using behavioral intention as the criterion. The findings revealed that the interactive effect of prior experience with attitude toward the act in predicting behavioral intention was not significant, and that the interactive effect of motivational set with normative belief in predicting behavioral intention was significant. In addition, a regression analysis of the BI-B relationship was performed in which the design variables were incorporated as predictors. Results showed that the inclusion of design variables aided the prediction of behavior significantly. Songer-Nocks' data were interpreted as being consistent with the overall Fishbein formulation, but as suggesting certain potentially specifiable limitations in predicting actual behavior from attitudinal components.     

Misconceptions revisited: A final comment

Although Songer-Nock's reanalysis of her data is consistent with our earlie comments, her reply indicates that some of her misconceptions persist.