Interesting conceptual framework for understanding performance on Gf tasks. However, it is Important to note that factor analysis studies have suggested a number of subprocesses that do not necessarily fit in this task-analysis based model.
Signatures of multiple processes contributing to fluid reasoning performance (
article link)
Ehsan Shokri-Kojoria and Daniel C. Krawczyk
A R T I C L E I N F O
Keywords: Fluid intelligence Individual differences Multi-process Raven's progressive matrices
A B S T R A C T
We aimed to achieve a better understanding of the cognitive processes of fluid reasoning (or fluid intelligence; Gf), the ability to reason in novel conditions. While fluid reasoning has often been considered a unitary con-struct, multiple cognitive processes are expected to affect fluid reasoning performance. Yet, the contribution of various cognitive processes in fluid reasoning performance remains under-explored. We hypothesized that in-dividual differences in fluid intelligence can be viewed as a composite of individual differences in performance in various processes of Gf. Change detection, rule verification, and rule generation were the three processes-of-interest that were additively recruited in a novel visuospatial reasoning task. We observed decreases in accuracy and increases in response time as the processing requirements increased across task conditions. Hierarchical multiple linear regression analyses showed that individual differences in the likelihood of success and speed of each of these processes, accounted for different aspects of individual differences in accuracy and response time in fluid reasoning performance, as measured by Raven's Progressive Matrices. Change detection was a significant contributor to performance in problems with higher visuospatial demand, however, rule verification and rule generation consistently contributed to performance for all problem types. Our findings support the position that individual differences in fluid intelligence emerge as a composite of performance on separable cognitive op-erations, with rule processing being important for differentiating performance on high difficulty problem.
Click on image to enlarge
- Posted using BlogPress from my iPad
Posted with Blogsy from my iPad mini 4
