Now you see it, now you don't: Relations between semantic activation and awareness

Implicit perception in patients with visual neglect: lexical specificity in repetition priming

Twelve patients with left-sided neglect, five patients with left-sided hemianopia, and 16 matched controls performed lexical decisions for foveally presented target words or pseudowords. Target words or pseudowords could be preceded by the same stimuli, presented for 150 ms as primes in either the left (LVF) or the right (RVF) visual hemifield. Primes in the RVF caused similar levels of priming across the three groups. In contrast, whereas primes in the affected LVF did not cause priming in hemianopic patients, patients with neglect showed highly significant priming from LVF primes, and the level of priming from these neglected primes was increased relative to normal control levels. This priming effect was attributed to lexical access, because no repetition priming was observed for pronounceable pseudowords. In striking contrast to the priming results, neglect patients' explicit recognition performance for these prime stimuli was at chance for the left hemifield, although it was near perfect for the right hemifield. This demonstrates that the observed priming effects in the left hemifield were indeed caused by stimuli that were not explicitly perceived.

Semantic priming without awareness: some methodological considerations and replications

Previous studies of subliminal semantic priming effects are considered. The distinction between subjective and objective detection thresholds proposed by Cheesman and Merikle (1984) is accepted, and it is concluded that no convincing evidence has been adduced to support or deny the existence of semantic priming effects at or below objective threshold. New criteria for the assessment of objective thresholds are proposed, and four experiments are reported in each of which pattern-masked primes were presented 10% below objective threshold. The first three experiments presented primes, mask, and targets binocularly; the fourth used dichoptic presentation. Data were analysed using both conventional F ratios and the quasi F ratio advocated by Clark (1973). Results showed that significant subliminal semantic priming effects occurred in each experiment when conventional F ratios were used, and significant effects occurred in all but the first experiment when quasi F was used. Binocular and dichoptic presentation of stimuli gave semantic priming effects whose magnitudes did not differ significantly. It is concluded that semantic priming effects can be obtained below objective detection threshold. It is also concluded that replication with different participants and different stimuli is preferable to the use of quasi F ratios in investigations where verbal stimuli are used.