Uncovering underlying processes of semantic priming by correlating item-level effects

Abstract

The current study examines the underlying processes of semantic priming using the largest priming database available (i.e., Semantic Priming Project, Hutchison et al. Behavior Research Methods, 45(4), 1099-1114, 2013. Specifically, it compares priming effects in two tasks: lexical decision and pronunciation. Task similarities were assessed at two different stimulus onset asynchronies (SOAs) (i.e., 200 and 1,200 ms) and for both primary and other associates. To evaluate how consistent priming is across these two tasks, item-level priming effects obtained in each task were correlated for each condition separately. The results revealed significant correlations at the short SOA for both primary and other associates. The correlations at the long SOA were significantly smaller and only reached significance when z-transformed response times were used. Furthermore, this pattern remained essentially the same when only asymmetric forward associates (e.g., panda-bear) were considered, suggesting that the cross-task stability at the short SOA was not merely caused by retrospective processes such as semantic matching. Instead, these findings provide evidence for a rapidly operating, item-based, relational characteristic such as spreading activation.

The (un)reliability of item-level semantic priming effects

“…Furthermore, if we only focused on Block 1 of Session 1, we would have a typical priming design, in the sense that no stimulus had been presented more than once at that juncture. This allowed us also to estimate the split-half reliability of item-level priming effects in the same manner as Heyman, Hutchison, and Storms (2016a) and Hutchison et al (2008).…”

“…As we mentioned before, some item-level priming effects are on average larger than others, such that, for instance, cat primes dog to a greater extent than does animal. However, these effects tend to vary considerably across participants (Heyman, Hutchison, & Storms, 2016a;Hutchison et al, 2008). For example, the reliability estimates for item-level priming effects obtained in the Semantic Priming Project (Hutchison et al, 2013) ranged from .08 to .35 (see Heyman, Hutchison, & Storms, 2016a, for more details).…”

“…However, these effects tend to vary considerably across participants (Heyman, Hutchison, & Storms, 2016a;Hutchison et al, 2008). For example, the reliability estimates for item-level priming effects obtained in the Semantic Priming Project (Hutchison et al, 2013) ranged from .08 to .35 (see Heyman, Hutchison, & Storms, 2016a, for more details). As such, the explanatory power of relatedness predictors is limited, because it is meaningless to predict noise.…”

“…Critically, the reliabilities of person-and item-level priming effects are not necessarily related to one another. One concerns the stability of an item characteristic, the other the stability of an individual trait (see Heyman, Hutchison, & Storms, 2016a, for further discussion). To be clear, the present study's focus was also not on the average semantic priming effect (i.e., the grand mean).…”

“…We predicted that items with a relatively large priming effect at T 1 should also show a large priming effect at T 2 . Given that Heyman, Hutchison, and Storms (2016a) found higher consistency across participants when the stimulus onset asynchrony (henceforth, SOA) was short (i.e., 200 ms) rather than long (i.e., 1,200 ms), we focused on conditions that should minimize the use of strategies (short SOA and low relatedness proportion-henceforth, RP). This study and others have suggested that there is at least some consistency.…”