Priming from novel masked stimuli depends on target set size

The semantic origin of unconscious priming: Behavioral and event-related potential evidence during category congruency priming from strongly and weakly related masked words

“…However, if the unconscious priming effects observed in our experiments mainly reflected either the involvement of actiontriggers for the semantic categories (Kiesel et al, 2006) and/or the mere congruency of prime and target stimulus-response mappings (a ''response priming" hypothesis), then there should not be any difference between strongly and weakly related word pairs, as both conditions are identical in terms of either category co-membership or S-R category congruency. So, a response priming account of congruency priming would predict equal priming in strongly and weakly related pairs.…”

“…For example, by using a masked congruency priming task Ortells et al (2013) found reliable priming effects from unpracticed prime words, only when they were followed by strongly related, but not by weakly related targets. These findings are difficult to explain in terms other than a semantic processing of masked words, as both strongly and weakly related pairs did not differ in terms of either prime-target orthographic overlap, response congruency (e.g., Wentura, 2000), stimulus-response mappings, or action-triggers for semantic categories (Kiesel et al, 2006). In contrast, if priming were based on response activation bypassing semantic prime processing, equal priming for strongly and weakly related pairs would be expected.…”

“…The lack of relation between both measures is common in the unconscious priming literature (e.g., Damian, 2001;Kiefer, 2002;Kiesel et al, 2006;Van den Bussche & Reynvoet, 2007). Such a finding could partly reflect small reliability in either task due to, for example, high individual differences in baseline RTs (e.g., Hutchison et al, 2008).…”

“…Of course, we do not exclude the possibility that nonsemantic mechanisms such as S-R associations or action triggering also contribute to masked congruency priming (e.g., Kiefer et al, 2015), given that experimental conditions other than in the present experiments are realized (e.g., masked primes also presented as visible targets, small categories). Kiesel et al (2006) have recently extended the action trigger hypothesis to explain unconscious priming from novel words and large target sets. They suggest that when many targets from broad categories (e.g., animals) are presented, subjects could form an action trigger for the expected semantic categories, instead of for the individual expected stimuli.…”

“…Such priming has been described as unconscious in nature when subjects are phenomenally unaware of the masked primes and/or they cannot identify them in a separate test of prime visibility. Evidence for reproducible unconscious congruency priming has been accumulated across a variety of categorization tasks, such as positive vs. negative valence judgments (De Houwer, Hermans, Rothermund, & Wentura, 2002;Kiefer, Sim, & Wentura, 2015;Klauer et al, 2007;Naccache et al, 2005), number classification (Dehaene et al, 1998;Naccache & Dehaene, 2001a, 2001b, size discrimination (Kiesel, Kunde, Pohl, & Hoffmann, 2006), gender classification Klauer et al, 2007), and category classification (Forster, Mohan, & Hector, 2003;Ortells, Daza, & Fox, 2003;Ortells, Frings, & Plaza-Ayllón, 2012;Ortells, Vellido, Daza, & Noguera, 2006;Van den Bussche & Reynvoet, 2007).…”