Perception and preference in short-term word priming.

Abstract

Responding optimally with unknown sources of evidence (ROUSE) is a theory of short-term priming applied to associative, orthographic-phonemic, and repetition priming. In our studies, perceptual identification is measured with two-alternative forced-choice testing. ROUSE assumes features activated by primes are confused with those activated by the target. A near-optimal decision discounts evidence arising from such shared features. Too little discounting explains the finding that primed words were preferred after passive viewing of primes. Too much discounting explains the findings of reverse preference after active processing of primes. These preference changes highlight the need to use paradigms (like the present ones) capable of separating preferential and perceptual components of priming. Evidence of enhanced perception was found only with associative priming and was very small in magnitude compared with preference effects.

Prime diagnosticity in short-term repetition priming: Is primed evidence discounted, even when it reliably indicates the correct answer?

“…reduction or reversal of the preference for primed words when comparing short and long duration primes. These discounting effects in comparing short and long duration primes were seen in all conditions of all experiments and replicate previous findings for nondiagnostic primes (e.g., Huber, Shiffrin, Lyle, & Quach, 2002;Huber et al, 2001;Weidemann et al, 2005).…”

“…This result is not surprising, given the fact that Weidemann et al (2005) also found priming effects when prime and target were presented in different colors. Indeed, even arguably more drastic differences between the prime and target appearances do not reduce these priming effects either: Huber et al (2001), for example, manipulated letter case between prime and target and found similar priming effects regardless of whether prime and target matched in case. These results suggest that the features used to identify the target are higher level abstract features rather than low-level aspects pertaining to the details of the presented stimulus (see Sanborn, Malmberg, & Shiffrin, 2004, for an interpretation of this finding as a masking effect).…”

“…This was revealed by higher accuracy when the target was primed and lower accuracy when the foil was primed. However, longer duration primes, or primes that were responded to, produced the opposite preference pattern, with lower accuracy when the target was primed but higher accuracy when the foil was primed (e.g., Huber et al, 2001;Weidemann et al, 2005).…”

“…A Model of Short-Term Priming Huber et al (2001) explained these findings with two distinct processes: (a) spatial and temporal confusions that cause some prime features to become mixed into the target percept (termed source confusion) and (b) discounting of evidence from percept features known to have been present in the prime(s) (and hence the prime [s] could have been the basis of the match). Source confusion introduces prime features into the target percept and evidence discounting works by reducing the impact of those features that are deemed likely to have been introduced by the prime.…”

“…Equation 5; .07 was a high enough value that it could be lowered sufficiently to produce the observed no-target effects). As in the modeling for previous experiments (e.g., Huber et al, 2001;Weidemann et al, 2005), we assumed that the probability that features were activated by the target and by noise was correctly estimated (␤Ј ϭ ␤, and ␥Ј ϭ ␥) with one exception: In Experiment 2, ␥ varied with target presence, but letting ␥Ј vary as well would imply that the system can distinguish between trials in which the target was present versus those in which a blank screen was presented instead. Therefore, we calculated ␥Ј for Experiment 2 as follows:…”