ERIC Number: ED049705
Record Type: RIE
Publication Date: 1971-Feb
Reference Count: 0
Generalized Learning Curves and their Ability and Personality Correlates.
Harris, Elizabeth L.; And Others
In learning experiments, individual trial performances are typically represented as a composite of two independent components. The first is associated with systematic variation, and the second is a residual component interpreted as error. From the systematic variation, mean trial performance is typically computed across trials to obtain a simple learning curve or function. Tucker has proposed representing systematic variations as two or more curves or components representing generalized learning functions. In an illustration of this method, Tucker reanalyzed data from a probability learning experiment performed by Gardner, finding in one condition that performances of individuals were best represented by three generalized learning functions. This study attempted to replicate Tucker's findings and to assess the relationship between cognitive and personality traits and individual scores associated with generalized learning functions. Specifically the study was concerned with (1) determining the number of learning functions required to determine individual variations in performance on a probability learning task, and (2) determining, through multiple regression analysis, a composite of cognitive and personality variables that are significant predictors of individual scores on each of the learning functions. (AF)
Publication Type: N/A
Education Level: N/A
Authoring Institution: American Educational Research Association, Washington, DC.
Note: Paper presented at the 55th Annual Meeting of the American Educational Research Association, New York, February, 1971