The suggestion that semantic activation can occur without conscious identification of the priming stimulus is still controversial. Many studies supporting such a contention, especially those where primes were auditorially presented, suffer from methodological shortcomings, frequently with regard to threshold measurement. In the study reported here 24 subjects underwent a considerably more rigorous thresholding procedure than has been usual, prior to engaging in a forced-choice sentence completion task. The results show that semantic priming operates when subjects were unable to detect the presence of primes and that phonological (but not semantic) priming operates when the primes were invariably detected but never correctly identified. The relevance of these qualitatively different effects of primes, as a function of the level at which they are presented, in discussed in the light of recent accounts of unconscious processing.
Detectionless processing with semantic activation? A footnote to Greenwald, Klinger, and Liu (1989)