UBC Theses and Dissertations
Reversal and nonreversal shifts in Indian and white children Kee, Herbert William
This study was designed to explore specific aspects of the relationship between language and cognition. Comparisons of a normal population with populations deficient in verbal ability provide information relevant to the qualification of this relationship. In this respect, B.C. Indian children were an appropriate group for comparison with normal white children since they are apparently deficient in verbal development. It was considered worthwhile to determine if there are cognitive differences between Indian and white children and if there are, to determine if these differences can be attributed to differences in verbal ability in the form of verbal mediation. Evidence of verbal mediation is assumed to be exemplified in the relatively greater ease of executing a R over a NR shift. In a 3 x 2 x 2 factorial design involving age (7,8,9), shift (R-NR), and ethnic group (Indian-white), it was hypothesized that there would be a significant interaction between shift and ethnic group. A total of sixty-seven Indian and fifty-one white children was initially tested. However, nineteen Indian and three white children failed to learn the first discrimination to criterion within the limit of one hundred trials. The difference between these proportions was highly significant. Analyses were conducted for the resulting self-selected sample of forty-eight Indian and forty-eight white children who succeeded in attaining the first criterion and who went on to the shift task. On original learning, there were no significant differences or interactions for this self-selected sample. On the shift, there was a significant main effect only for the shift factor, with the R shift performance being superior to MR shift performance for both ethnic groups. There were no differences between Indians and whites in overall performance or in the relative difficulty of R and NR shifts. Supplementary analyses were performed to explore other possible differences. It was found that the white children were relatively consistent in the speed with which they learned both the original discrimination and shift while, in contrast, the Indian children were not. Those Indian children who were "fast" in original learning became "slow" on the shift, whereas those who were"slow" in original learning became "fast" on the shift. On the basis of post-experimental card sort and verbalization tests, it was also found that the shape dimension was more salient than the size dimension and that Indian children were not as successful in giving an appropriate overt label to the triangle concept. The specific hypothesis that there would be a significant interaction between shift and ethnic group was not supported. However, in general, the results from the supplementary analyses and the fact that significantly more Indian than white children failed to reach the first criterion suggested that there wore cognitive differences between Indian and white children. There was no specific evidence to support a mediational deficiency interpretation of these differences.
Item Citations and Data