Learning Science: Some Insights from Cognitive Science

Theories of teaching and learning, including those associated with constructivism, often make no overt reference to an underlying assumption that they make; that is, human cognition depends on domain-free, general-purpose processing by the brain. This assumption is shown to be incompatible with evidence from studies of children's early learning. Rather, cognition is modular in nature, and often domain-specific. Recognition of modularity requires a re-evaluation of some aspects of current accounts of learning science. Especially, children's ideas in science are sometimes triggered rather than learned. It is in the nature of triggered conceptual structures that they are not necessarily expressible in language, and that they may not be susceptible to change by later learning.