Abstract

Even though the performance in the visual recognition memory tested with FTII proved to be a valid predictor for the later intelligence, the low internal reliability as well as low test-retest reliability have been problematic for the utility of FTII. This study examined the possible explanation for the fluctuation of the performance in FTII. 52 infants between 67 and 92 weeks of conceptional age were tested with the newly developed computerized version of FTII and their temperaments were measured using Infant Characteristic Questionnaire (ICQ) which was originally developed by Bates (1979). In the separate analyses of 16 infants' performance both in this computerized version of the FTII and the original FTII revealed the concurrent validity of it. These 16 infants between 67 and 92 weeks of age were tested twice within a week with the computerized version (tested first) as well as the original FTII. Subjects were from both the full-term (N=40) and premature group (N=12). But there was no age or group differences in both the performance of FTII and the ratings of the ICQ. Regression analyses revealed that among 4 dimensions of temperament such as difficulty, adaptability, dullness and unpredictability the dimension of dullness significantly predicted the novelty preference (r=-.39, p<.01). It explained 15. 3% of the variance of visual recognition memory tested with FTII. The dimension of difficulty which has been proved to be the most important temperamental aspect was correlated only marginally with the performance of FTII, r=-.23, p<.10. These results suggested that the performance of FTII can be influenced by the temperament of infants and the complex interaction of the temperament and the situational variability may contribute the variability of performance in the infant intelligence tests such as FTII.