Psychology Department

Health Psychology Home Page

Papers written by students providing scientific reviews of topics related to health and well being

  HomeWeight LossAlternative Therapy | Supplements | Eating Disorders | Fitness | Links | Self-Assessment | About this Page |


Is Breastfeeding the Smart Decision? The Effect of Breastfeeding on Children’s Intelligence Levels

Image:Infant looking at shiny object.jpg



Joe Knadler

September 24, 2007



            In a time when admission to the nation’s elite colleges is growing more competitive each year, many parents are anxious to start their children early on a path to academic success. The booming early infant education industry, demonstrated by the success of such products as the Baby Einstein series, seems to indicate that parents are willing to make substantial investments to improve their baby’s intelligence.  A quick internet search will reveal hundreds of sites promoting different regimens for accomplishing this goal, from a Mensa-endorsed vocabulary and language program ( to a fish-based diet for toddlers   (

            While some of these techniques may have a positive effect on intelligence levels, few have been as extensively debated as breastfeeding.  In 1929, the first article proposing a relation between breastfeeding in infancy and higher levels of intelligence was published (Hoefer and Hardy, 1929). Over the last eighty years, hundreds of studies have examined the validity of this claim, yielding both positive and negative results.  Reviewing the body of literature on the subject, nearly all researchers agree that breastfeeding has a clear positive effect on the health of both mother and child; however, the effect of breastfeeding on intelligence levels appears to be much weaker than initially claimed, if existent at all.  In small for gestational age or premature infants, however, the effect of breastfeeding on intelligence levels may be more significant (Agostoni, 2005).



How Does it Work?

1) Breastfeeding Increases Intelligence Levels Through the Action of DHA and PUFAs

            While many correlational studies have claimed to demonstrate linkage between breastfeeding and intelligence, few propose a causative mechanism by which this effect may occur.  Where an explanation is provided, many of the beneficial health effects of breastfeeding on infants have been attributed to the presence of large amounts of long-chain polyunsaturated fatty acids (PUFAs) in breast milk compared to infant formula (Schack-Nielsen and Michaelsen, 2007).  Docosahexaenoic acid (DHA), one of these long chain PUFAs, has been most intensely investigated, since it has been demonstrated that this molecule is taken up by cell membranes of the central nervous system (Lauritzen, Hansen, Jorgensen, & Michaelsen, 2001).  Through some as yet undetermined mechanism, DHA may act to promote neural and cognitive development, thus accounting for the increased intelligence observed in breastfed infants.  One study by Jensen et al. (2005) tested this hypothesis by supplementing the diet of new mothers with DHA or a placebo for the first four months after birth using a double-blind experimental design.  While mother’s milk is already high in DHA, further increase in DHA levels in milk through supplementation of maternal diet was expected to result in higher intelligence levels in the breastfeeding infants. By examining the levels of DHA in both the mother’s milk and the plasma of their breastfeeding infants, they determined that this supplementation had effectively increased the level of DHA received by the infants in the DHA supplementation condition.  The researchers then examined whether increased levels of DHA were correlated with better performance on cognitive tests and visual acuity.  While no significant effects were found at four, eight, or twelve months of age, infants receiving higher levels of DHA performed better on psychomotor tests at thirty months of age.

            Other research studies have attempted to examine whether supplementation of infant formula with DHA and other PUFAs can reduce intelligence differences between breastfed and formula-fed infants.  In a 2007 study, Birch et al. examined visual acuity and intelligence differences between a nonrandomized sample of breast-fed children and randomized samples of children receiving control formula, children receiving control formula supplemented with DHA, and children receiving control formula supplemented with DHA and arachadonic acid (ARA). Formula-fed infants were randomly assigned to one of the three different formula conditions and received only this formula for the first seventeen weeks after birth.  Appraisal of cognitive abilities using the Wechsler Preschool and Primary Scale of Intelligence revealed no significant differences in overall IQ score at four years of age among the different conditions; however, children receiving control formula or formula supplemented with only DHA showed a significantly lower verbal IQ score compared to breastfed children.  Children receiving formula supplemented with both DHA and ARA were not different from breastfed children.  The fact that supplementation of formula with DHA and ARA reduced the difference between breastfed and formula-fed infants suggests that some of the effect of breastfeeding on intelligence levels may be mediated by these molecules and other PUFAs.




2) Breastfeeding Increases Intelligence Levels by Physical, Psychological, and Maternal Actions

            Other research has suggested that the composition of breast milk is not the primary cause of the observed increase in intelligence, but rather, the act of breastfeeding has secondary effects that may promote increased intelligence. The behavior of breastfeeding may promote cognitive development due to the physical and psychological contact between mother and infant that occurs in this process (Schack-Nielsen and Michaelsen, 2007).  Several studies have also demonstrated that breastfeeding increases maternal levels of the hormones oxytocin and prolactin, which promote better memory and cognitive performance in the mother (Silber et al., 1990; Heinrichs et al., 2004).  Such an effect may lead to increased attention and focus on the infant, lower levels of depression, and more affectionate physical contact with the infant, which may all be associated with a greater concentration on promoting the cognitive development of the newborn.  These various mechanisms are summarized in the figure below from a review on the biological effects of breastfeeding (Schack-Nielsen and Michaelsen, 2007).

(from Schack-Nielsen and Michaelsen, 2007; LCPUFA= Long chain polyunsaturated fatty acids)


Who is Advocating Breastfeeding?

            While few if any organizations are recommending breastfeeding solely for the purpose of increasing a child’s intelligence levels, breastfeeding meets with nearly universal approval from both physicians and government health agencies.  The American Academy of Pediatrics (, the United States Surgeon General ( ), and the World Health Organization ( ) all endorse breastfeeding, especially during early infancy, for the proper nutrition, growth, and health of a child. The U.S. Surgeon General recommends breastfeeding exclusively for the first six months of life, and strongly recommends continuing until at least one year of age ( The health benefits of breastfeeding, especially in the area of improved immune functioning, have been well-documented (Schack-Nielsen and Michaelsen, 2007).  Additional benefits for the child include ensuring proper growth and nutrition ( The numerous positive benefits for the mother include decreased monetary costs compared to formula, loss of weight gained during pregnancy, better bonding with the newborn, and the possible reduction of risk for breast and ovarian cancers (


Scientific Research and Reviews

            Starting in 1929 with the publication of the first study by Hoefer and Hardy indicating that breastfeeding might cause higher levels of intelligence, a large body of scientific research has been amassed to test this hypothesis.  Due to ethical constraints, true experimental research has been difficult to carry out in this field.  Faced with the overwhelming evidence supporting the positive effect that breastfeeding can have for both child and mother, randomly assigning participants to a non-breastfeeding condition would be morally suspect.  Research must instead rely on women who voluntarily choose not to breastfeed, a fact which introduces a large number of potential confounding variables, since certain maternal characteristics are associated with an increased likelihood of breastfeeding, most notably maternal intelligence (Der, Batty, and Deary, 2006).

            A meta-analysis conducted by Anderson, Johnston, and Remley (1999) examined twenty studies comparing the levels of intelligence of formula-fed and breastfed infants.  Only eleven studies fit the criteria of controlling for at least five of the fifteen associated covariables (e.g. maternal intelligence, socioeconomic status, race, maternal age) and reported scores that were unadjusted, and adjusted in response to these possible confounds.  After intensive statistical analyses, the authors concluded that a significant difference between formula-fed and breastfed infants was observed in cognitive performance.  The unadjusted difference of 5.32 points on an intelligence quotient scale was determined from the pooled study data.  When adjusted for applicable confounds, this difference decreased to 3.16 points, but remained significant.  A larger effect was observed when only preterm babies were considered; breastfed premature babies showed a 5.18 point increase over their formula-fed peers.  These results support the idea that breastfeeding can result in a small benefit on a child’s intelligence, an effect which is magnified in premature children.  The analysis supported the observed trend that most research has indicated a small increase in intelligence level among those who are breastfed.

            A more recent review by Jain, Concato, and Leventhal (2002) has challenged the validity of this previous meta-analysis based on methodological flaws.  These researchers claim that in the previously-mentioned study, the authors failed to account for the variation in strength of methodology of the studies included in their meta-analysis.  Jain, Concato, and Leventhal (2002) examined forty studies conducted between 1929 and 2001 regarding the effect of breastfeeding on intelligence.  Three types of studies were examined. The first type was birth cohort design, where the individuals were recruited at birth and their feeding history and cognitive outcome followed over subsequent years.  The second type was school registry cohort studies, where participants were recruited from school lists and asked retrospectively about their feeding habits as infants. The third grouping of studies was case control design, where individuals with poor cognitive performance were identified and their feeding habits were retrospectively analyzed and compared to a control group. Of the forty studies examined, twenty-seven concluded that breastfeeding promoted higher levels of intelligence (sixty-eight percent). Upon closer analysis, however, the authors concluded that all but two of the forty studies had methodological flaws that made their results suspect.   The major issues with many of the studies were that the researchers failed to adequately control for socioeconomic status, maternal intelligence, and child stimulation.

            The results of the two studies the authors found to be adequate were mixed.  One claimed an effect of breastfeeding on intelligence and the other provided no conclusive evidence.  While the authors admit that the body of data supporting a positive effect of breastfeeding on intelligence is large, they claim that the higher the methodological quality of the study, the less persuasive the obtained results are. 

            An even more recent review of the literature on breastfeeding and cognitive development attempted to more completely account for the affect of the potential confounding variables described in previous studies.  Der, Batty, and Deary (2006) examined data from the 1979 US National Longitudinal Survey of Youth.  In order to eliminate the effects of confounding variables, they also analyzed the child’s home environment, demographic characteristics, and maternal characteristics.  In looking at associations between the possible confounds and breastfeeding behavior, the authors found that higher maternal IQ, higher maternal education, older maternal age, higher socioeconomic level, and more stimulating and supportive home environments were all positively correlated with breastfeeding.  Since all of these characteristics might also be expected to positively influence a child’s cognitive development, they represent strong potential confounds in any associations between intelligence and breastfeeding.  Maternal intelligence was the strongest predictor of breastfeeding, with those mothers more than one standard deviation above average intelligence being twice as likely to breastfeed their children.  Since intelligence is a moderately heritable trait, this potential confound can go a long way in explaining any observed increase in intelligence in breastfed children.

            Unadjusted scores revealed an approximately four point increase in mental ability on the IQ scale for breastfed children compared to formula-fed children, a statistically significant difference comparable to those observed in many other studies.  However, adjusting for confounding variables of maternal IQ, maternal education, maternal age, family poverty, home environment stimulation level, and birth order, the effect decreased to only a 0.52 point increase in IQ score of breastfed children, which was no longer statistically significant. 

In order to more completely account for confounding variables, the authors also undertook a sibling-pairs analysis where siblings differed in the duration of breast feeding or breastfeeding status.  In these sibling pairs, when the breastfed child was compared to the formula-fed child, no significant difference in intelligence was observed. Likewise, when siblings were breastfed for different periods of time, no difference in intelligence levels were seen. The fact that no significant differences in intelligence emerged in these sibling comparisons suggests that maternal and environmental confounding variables account for virtually all of the observed differences in intelligence between breastfed and formula-fed infants.

            In addition to this statistical analysis, Der, Batty, and Deary (2006) conducted a meta-analysis of twelve previous studies on breastfeeding and intelligence.  Again, they found that few of the studies had adequately controlled for all possible confounds.  While the effect size for studies that controlled only for maternal intelligence revealed a 3.37 point increase in IQ score for breastfed children, when all confounds were controlled for, the effect decreased to a far from significant result of only a 0.16 point increase in IQ score.  The authors conclude that previous studies have falsely attributed increased intelligence to breastfeeding by failing to control for important maternal and environmental confounds.  In reality, breastfeeding has a very small or nonexistent effect on increasing intelligence in children; maternal IQ is a much stronger predictor of intelligence level than breastfeeding status.

            The body of scientific literature on the subject of breastfeeding and intelligence is full of conflicting reports, as evidenced in the previously cited studies.  While few researchers can agree on whether breastfeeding itself is responsible for increasing intelligence levels in children, it is readily apparent that any difference between intelligence levels of breastfed and formula-fed infants is small in size, at most only a few IQ points.  Careful control for confounding variables further decreases these differences and reveals that many other factors are much more important in determining a child’s intelligence level than breastfeeding.


The Special Case of Premature and Underweight Babies

            The previously detailed studies have all focused on the cognitive development of full-term babies.  When the effect of breastfeeding on premature and underweight infants is examined, a clearer difference results.  Slykerman et al. (2005) conducted a prospective cohort study examining the intelligence quotient of children at 3.5 years of age who had been breastfed or formula-fed as infants. While no difference in intelligence levels was found when looking at the whole group of breastfed infants compared to formula-fed infants, small for gestational age children (underweight at birth) showed intelligence levels that were correlated with feeding style.  Breastfed infants showed an increase of 6.0 IQ points over formula-fed infants among underweight children.  Other studies have cited increases of at least 5.1 IQ points in premature babies who are breastfed rather than formula-fed (Anderson et al., 1999).  These results are not surprising, since premature babies show a greater benefit of breastfeeding in a number of other domains, including growth, development, and immune strength (Schack-Nielsen and Michaelsen, 2007). 


The Bottom Line

            While much research has suggested a link between breastfeeding and higher intelligence levels in later life, when the data are reevaluated to account for confounding variables such as maternal IQ, maternal education, and home environment, the effect size greatly decreases.  Maternal IQ alone accounts for up to seventy-five percent of the observed correlation between breastfeeding and increased intelligence (Der, Batty and Deary, 2006). While the benefits of breastfeeding are numerous, including a stronger bond between mother and child and better overall infant health, increased intelligence may not be one of them (Schack-Nielsen and Michaelsen, 2007).  Further research will need to be conducted in order to better understand if breastfeeding has a small effect on intelligence levels. While all mothers should consider breastfeeding for both the health of themselves and their child, their motivation to do so should not be based upon the hope that doing so will increase their child’s intelligence level. 


















Agostoni, C. (2005). Small-for-gestational-age infants need dietary quality more than

quantity for their development: The role of human milk. Acta Paediatrica, 94(7), 827-829.

American Association of Pediatrics. Breastfeeding. Retrieved September 22, 2007 from

Anderson, J.W., Johnston, B.M., & Remley, D.T. (1999). Breast-feeding and cognitive development: a meta-analysis. Am. Journal of Clinical Nutrition, 70(4), 525-535.

Birch, E.E., Garfield, S., Castaneda, Y., Hughbanks-Wheaton, D., Uauy, R., and Hoffman, D. (2007). Visual acuity and cognitive outcomes at 4 years of age in a double-blind, randomized trial of long-chain polyunsaturated fatty acid-supplemented infant formula. Early Human Development, 83, 279-284.

Cone, M. (2005). Calculated, risk is worth benefit of eating fish. Retrieved September 22, 2007 from

Der G., Batty G.D., & Deary I.J. (2006). Effect of breast feeding on intelligence in children: prospective study, sibling pairs analysis, and meta-analysis. BMJ, 333(7575), 945. E. pub 10.1136/bmj.38978.699583.55.

Heinrichs M., Meinlschmidt G., Wippich W., Ehlert U., & Hellhammer D.H. (2004). Selective amnesic effects of oxytocin on human memory. Physiol Behav, 83, 31-38.

Hoefer C, Hardy M. (1929). Later development of breast fed and artificially fed infants. JAMA

92, 615-620.

Jain, A., Concato, J., & Leventhal, J.M. (2002). How good is the evidence linking breastfeeding and intelligence? Pediatrics, 109, 1044-1053.

Jensen C.L., Voigt R.G., Prager T.C., Zou Y.L., Fraley J.K., Rozelle J.C., Turcich M.R., Llorente A.M., Anderson R.E., & Heird W.C. (2005). Effects of maternal docosahexaenoic acid intake on visual function and neurodevelopment in breastfed term infants. American Journal of Clinical Nutrition, 82(1), 125-132.

Lauritzen L., Hansen H.S., Jorgensen M.H., & Michaelsen K.F. (2001). The essentiality of long chain n-3 fatty acids in relation to development and function of the brain and retina. Prog Lipid Res, 40,1–94.

Madsen, P. (1998). Increasing Your Baby’s Intelligence. Retrieved September 22, 2007, from

Schack-Nielsen, L. & Michaelsen, K.F. (2007). Advances in our understanding of the biology of human milk and its effects on the offspring. The Journal of Nutrition, 137(2), 503S-510S.

Silber M., Almkvist O., Larsson B., & Uvnas-Moberg K. (1990). Temporary peripartal impairment in memory and attention and its possible relation to oxytocin concentration. Life Science, 47, 57–65.

Slykerman R.F., Thompson J.M., Becroft D.M., Robinson E., Pryor J.E., Clark P.M., Wild C.J., & Mitchell E.A. (2005). Breastfeeding and intelligence of preschool children. Acta Paediatrica, 94(7), 832-837.

United States Department of Health and Human Services. (2007). Breastfeeding—best for baby. Best for mom. Retrieved September 22, 2007 from

World Health Organization. (2007). Breastfeeding. Retrieved September 22, 2007 from



Psychology Department

The Health Psychology Home Page is produced and maintained by David Schlundt, PhD.

Vanderbilt Homepage | Introduction to Vanderbilt | Admissions | Colleges & Schools | Research Centers | News & Media Information | People at Vanderbilt | Libraries |Vanderbilt Register | Medical Center 

  Return to the Health Psychology Home Page
  Send E-mail comments or questions to Dr. Schlundt