Experimental Programming of Life Histories

Abstract

Four experimental variables-whether or not the mother had been handled during her infancy, the presence or absence of handling of the pup, whether the young were born and reared in a maternity cage or a free environment prior to weaning, and whether the young were reared in a laboratory cage or a free environment between weaning and 42 days of age-were combined in a 24 factorial design (N = 6 Purdue-Wistar rats per group). Starting at 220 days of age the groups were given an extensive battery of tests. Analyses of variance extablished that there were many significant treatment (i.e., main effects and interactions) among the dependent variables. These significant may be attributed to the four independent variables, and the contribution of genetic variance to the treatment may be considered to be of a negligible amount. Intercorrelations were then computed among all the dependent variables, using the group mean as the basic score (i.e., N = 16 for each correlation coefficient). The intercorrelation matrix was factor analyzed. Three clearly identifiable factors were extracted. They were: emotional reactivity, field exploration, and consumption-elimination. A fourth factor, which could not be readily identified, was most heavily weighted with two measures of avoidance learning. The results establish that significant differences can be created by experimental means independent of any contribution from genetic variance. These experimentally created individual are highly stable, persisting for a period of at least 233 days. The isolation of a factor of emotional reactivity supports DENENBERG'S (1964) position that one of the major effects of early experience is to modify emotionality. The finding that emotionality and exploration are two independent factors has several theoretical implications.