Researchers have considered your hypothesis. There have been many attempts to eliminate confounding variables in these studies (this doesn't mean that they were

Researchers have considered your hypothesis. There have been many attempts to eliminate confounding variables in these studies (this doesn't mean that they were necessarily successful in doing so, but they did think about it). Often they attempt to control for differences in parenting styles, which should cover how much time they spend interacting with their child. Also until 6-10 months, babies can't hold their own bottles, so someone is probably still interacting with them (unless you prop up the bottle and leave the baby alone, which is possible but not recommended).