What is the role of consciousness in language processing? Unconscious priming experiments show that words can prime other words with related meanings (cat – dog), and these priming effects are assumed to reflect the activation of conceptual knowledge in semantic memory. Alternatively, however, unconscious priming effects could reflect predictive relationships between the words’ forms, since words that are semantically related are also statistically related in language use. Therefore, unconscious “semantic” priming effects could be due to relationships between words’ forms mimicking conceptual relationships, as in Searle’s Chinese Room thought experiment. To distinguish wordform-based and semantics-based accounts of priming we conducted an experiment in which temporal words (e.g., earlier, later) were preceded by spatial words that were processed either consciously or unconsciously. Time is typically conceptualized as a spatial continuum extending along either the sagittal (front-back) or the lateral (left-right) axis, but only the sagittal space-time mapping is encoded in language (e.g. the future is ahead, not to the right). Results showed that temporal words were primed both by sagittal words (back, front) and lateral words (left, right) when primes were perceived consciously, as predicted by both wordformbased and semantics-based accounts. Yet, only sagittal words produced an unconscious priming effect, as predicted by the wordform-based account. Unconscious word processing appears to be limited to relationships between words’ forms, and consciousness may be needed to activate words’ meanings.
