Evidence is accumulating that a dietary supply of long-chain polyunsaturated fatty acids (LC-PUFAs) enhances the development of attention and efficient information processing in infants. However, it is uncertain whether LC-PUFAs in infancy influence cognitive development in later childhood. The objective was to determine the effects of dietary LC-PUFAs in infancy on measures of cognitive function at age 6 y. Infants were randomly assigned to receive formula containing either docosahexaenoic acid and arachidonic acid or no LCPUFAs for a period of 4 mo. A reference breastfed group was also included. In a follow-up conducted at age 6 y, children received assessments of intelligence quotient (IQ), attention control (Day-Night Test), and speed of processing on the Matching Familiar Figures Test (MFFT). At follow-up there were 71 children in the LC-PUFA group, 76 in the control group, and 88 in the breastfed group. The formula groups did not differ on measures of Full-Scale IQ (LCPUFA mean = 98.0; control mean = 100.9) or attention control (LCPUFA mean = 12.7; control mean = 12.8). MFFT error scores were the same for both formula groups, but when making correct responses, the LC-PUFA group was significantly faster (mean = 6.2 s) than the control group [mean = 7.8 s; F(1, 131) = 6.09, P = 0.015]. IQ scores of children who were fed a formula containing either LC-PUFAs or no LC-PUFAs did not differ at age 6 y. However, children who received LC-PUFAs were faster at processing information compared with children who received unsupplemented formula. Variation in the dietary supply of LC-PUFAs in the first months of life may have long-term consequences for the development of some cognitive functions in later childhood.