Insights into the origins of knowledge from the cognitive neuroscience of blindness

Abstract

Children learn about the world through senses such as touch, smell, vision, and audition, but they conceive of the world in terms of objects, events, agents, and their mental states. A fundamental question in cognitive science is how nature and nurture contribute to the development of such conceptual categories. What innate mechanisms do children bring to the learning problem? How does experience contribute to development? In this article we discuss insights into these longstanding questions from cognitive neuroscience studies of blindness. Despite drastically different sensory experiences, behavioural and neuroscientific work suggests that blind children acquire typical concepts of objects, actions, and mental states. Blind people think and talk about these categories in ways that are similar to sighted people. Neuroimaging reveals that blind people make such judgements relying on the same neural mechanisms as sighted people. One way to interpret these findings is that neurocognitive development is largely hardwired, and so differences in experience have little consequence. Contrary to this interpretation, neuroimaging studies also show that blindness profoundly reorganizes the visual system. Most strikingly, developmental blindness enables “visual” circuits to participate in high-level cognitive functions, including language processing. Thus, blindness qualitatively changes sensory representations, but leaves conceptual representations largely unchanged. The effect of sensory experience on concepts is modest, despite the brain's potential for neuroplasticity.

Keywords:

• Conceptual development
• Innateness
• Experience
• Blindness
• Concepts

Acknowledgments

We are grateful to the blind individuals, parents of blind children and the blind community for enabling the research described in this article. We would like to thank Lindsay Yazzolino for sharing her insightful reflections and Amy Skerry for comments on an earlier version of this draft. We would also like to thank the Packard Foundation and the NeuroDiscovery fund for their generous support of this research.