Symbolic communication in the form of language underlies our unique ability to reason—or so the conventional wisdom holds. A new study published today in Science, though, suggests our capacity to reason logically may not actually depend on language, at least not fully. The findings show babies still too young to speak can reason and make rational deductions.

The authors—a team hailing from several European institutions—studied infants aged 12 and 19 months, when language learning and speech production has just begun but before complex mastery has been achieved. The children had to inspect distinct objects repeatedly—such as a dinosaur and a flower. The items were initially hidden behind a black wall. In one set of experiments the animation would show a cup scooping up the dinosaur. Half of the time, the barrier would then be removed to reveal, as expected, the remaining flower. In the rest of the instances, though, the wall would disappear and a second dinosaur would be there

The children deduced in these latter occurrences that something was not quite right, even though they were unable to articulate in words what was wrong. Eye-tracking—a commonly used technique to gauge mental abilities in preverbal children and apes—showed infants stared significantly longer at scenes where the unexpected object appeared behind the barrier, suggesting they were confused by the reveal. “Our results indicate that the acquisition of logical vocabulary might not be the source of the most fundamental logical building blocks in the mind,” says lead study author Nicoló Cesana-Arlotti, a postdoctoral fellow in the Department Psychological and Brain Sciences at Johns Hopkins University. A major component of human logic, he notes, relates to thinking about alternative possibilities and eliminating inconsistent ones: Does the dinosaur sit behind the barrier or does the flower? In a formal logic this is called a disjunctive syllogism: A or B; not if A, therefore B. (A syllogism is a conclusion derived from two distinct premises.

As part of their study, Cesana-Arlotti and his colleagues also reported infants’ pupils dilated when watching animations featuring illogical outcomes. This is known to occur in adults tasked with logic problems and provides more evidence babies are aware of the way things “should” be. “Their approach of using multiple trial types is very strong,” says Johns Hopkins psychologist and reason researcher Justin Halberda, who was not involved in the study but wrote an accompanying analysis in Science about the new paper. “I think many people would say that most of their reasoning happens when they are silently talking to themselves in their heads. What this new study reveals is that preverbal infants are also working through this same type of serial reasoning, and doing so before robust language abilities have been mastered.”

Cesana-Arlotti acknowledges his findings do not negate the importance of language and symbolic communication to human brain development, and to our evolutionary backstory. Yet the new research suggests that perhaps it is not entirely necessary to shape the brain’s logical reasoning capacities. He plans further work studying how preverbal logic might still differ from reasoning abilities that emerge once language comes along, as language may open additional reasoning abilities unavailable to the speechless brain. He also hopes to explore more deeply the mental development of young children. “Our research aims to investigate the earliest foundations of our ability to reason logically,” he says, “a major basis for learning, creativity and flexibility in the human mind.”

“To our knowledge, nobody has ever directly documented logical reasoning in 12-month-old infants before,” he adds. “The exploration of the initial state of logic in the mind is a very exciting enterprise.”