Zajonc and his coauthors have presented three confluence models that attempt to explain variations in intellectual development by the size and spacing of an individual's sibship and, most recently, by the age of the individual at testing. A closer look at two of these models revealed several major problems of internal consistency and external correspondence to data. The use of the child spacing variable in both models was particularly suspect. In addition, the most recent confluence model appears to have little applicability to within-families data due to the model's inability to account for individual differences in intellectual growth functions among siblings. The past use of between-families data for substantiating within-families models is also discussed. Finally, new between-and within-families data showed essentially no relationship between sibling spacing and intellectual development.

Studies are reported which show that concrete and abstract words of equal objective frequency (based on available word counts) are not perceived as being equal. The abstract word has greater perceived frequency than the concrete word. The judged variety of contexts in which a word appears correlates very highly with perceived frequency. The results have relevance to the design of learning studies in which concrete and abstract words are used, and also to the interpretation of such experiments.The studies to be reported here were prompted by the results of a minor experiment carried out by the first author as a part of the work in a graduate course in psychophysics. The S's task was to choose the word of a pair which had the higher background frequency. It was observed that when pairs contained an abstract word and a concrete word of equal background frequency as indexed by the Thorndike and Lorge (T-l) (1944) tables, the abstract words were chosen for an overwhelming majority of the pairs. It was as if the Ss had a strong bias for selecting the abstract word in a forced-choice situation in which the two words were equal on the characteristic being scaled (frequency). Such a finding, if confirmed, has several implications. These implications will be discussed later. The purpose of the first experiment was to determine if the finding could be replicated. EXPERIMENT I Method MaterialsA total of 595 nouns was selected initially from the list provided by Paivio, Yuille, and Madigan (1968). For the present study, concrete words were defined as those having C ratings greater than 4.0 on the 7-point scale, and abstract words were those having ratings less than 4.0. The mean value of the concrete words was 6.46, with a range of 4.26 to 7.00. The mean for the abstract words was 2.50, the range being 1.42 to 3.77.Eight levels of I-L frequency were selected for testing: 2, 10,20, 25,30,40, A, and AA.The Ss were tested by the forced-choice procedure. Two types of pairs were constructed, equivalent pairs and variable pairs. An equivalent pair consisted of a concrete word and an abstract word of equal T-l frequency, equal number of letters, and equal number of syllables. There were fewer abstract than concrete words available, particularly in the middle frequency levels. Therefore, in forming the pairs, the abstract words were selected first, being drawn randomly from the eight frequency levels.The concrete words were then grouped according to frequency level, number of letters, and number of syllables. From these groupings, a word was chosen randomly to correspond to a previously selected abstract word. Five such equivalent pairs were formed for each of the eight frequency levels. The frequency values for the two words in an equivalent pair were not always identical, particularly at the higher frequency levels, 56 but when a difference was present, it was small and, across pairs, unsystematic. The mean values at the eight frequency levels were very close to those specified above. The frequencies for A and AA words were dete...

This study assessed participant satisfaction with two interpretation formats and the effects of taking the RELATionship Evaluation (RELATE) on single young adults' premarital relationships. Thirty-nine engaged or seriously dating couples were assigned to one of three groups: (a) those who took RELATE and interpreted the results themselves, (b) those who took RELATE and participated in an interpretation session with a therapist, or (c) a control group. Results showed that taking RELATE with therapist assistance had a significant positive effect on perceived relationship satisfaction, commitment, opinions about marriage, feelings about marriage, and readiness for marriage. Positive effects also included increased awareness of strengths and challenges, improved couple communication, and the expectation of the prevention of future relationship problems. Taking RELATE without therapist assistance produced a small initial drop in relationship satisfaction followed by a marked improvement over time. Both genders approved of two interpretation formats-self-interpretation and therapist-assisted interpretation-with males slightly preferring therapist assistance. These results add to the literature on the usefulness of brief assessment techniques as effective interventions with premarital couples.

Unidirectional associates were used in a test of the encoding specificity principle with single items. Strong preexperimental associates were effective retrieval cues even when encoding conditions were not conducive to the establishment of a target-cue association. While the strength of the association from the target to the cue may have been involved in the cuing effect, the most important consideration was the presence of an association from the cue to the target. A comparison of the results from Experiments 1 and 2 suggested that the presence of weak cues on the test reduced the effectiveness of strong cues. A third experiment confirmed this finding. The generality of these findings to the encoding specificity principle was also discussed.

The focus of this report was the degree of independence of memory attributes in determining performance. A demonstration experiment indicated that memory can carry an associative attribute and a frequency attribute simultaneously with little interaction between the two, that is, they were independent in their influence on performance. The discussion centered on attribute control as a function of perceived task demands, on the distinction between attribute composition and attribute utilization, and on certain problems inherent in the analysis of memory and the corresponding problems in theory construction.