A Mouse IQ Test

See allHide authors and affiliations

Science  08 Aug 2003:
Vol. 301, Issue 5634, pp. 735
DOI: 10.1126/science.301.5634.735d

General cognitive ability has been the focus of heated debate over its measurement, its significance, and the degree to which innate and learned components contribute. Proficiency on standardized functional tests has been incorporated into a metric commonly referred to as the intelligence quotient (IQ). Matzel et al. have developed a similar performance-based approach to quantifying learning ability and applied it to a sample of 56 genetically heterogeneous mice. They chose five dissimilar tasks (whose acquisition would not transfer between tasks) in order to impose demands on sensory and motor skills and on the processing of information and motivation. The relative rankings of individual mice across the battery of tests were consistent (and replicated on a second test series); that is, an efficient learner of the Lashley maze also learned to associate tone and electrical shock quickly. Compilation of the data yielded a bell-shaped distribution in learning ability, and a principal component analysis revealed a factor that accounted for 38% of the variance between individual mice. This is comparable to the contribution of general intelligence (the g factor) to the range of human performance on IQ tests (which probe verbal, spatial, and other domains). — GJC

J. Neurosci. 23, 6423 (2003).

Stay Connected to Science

Navigate This Article