Semantic priming and identification of near threshold primes in a lexical decision task

Three experiments examined both the impact of semantic analysis of 50-msec, masked visual primes on a target response and the impact of semantic analysis of the target on a prime response. The first two experiments used a prime-target interval of 1000 msec. In Experiment 1, subjects reported the identity of each prime: (a) after a lexical decision about the target, (b) both before and after a lexical decision, or (c) after a target detection response. Prime report after both types of target response showed retroactive priming in which report was facilitated by related targets and inhibited by unrelated targets. Analyses of lexical decision latency and accuracy conditionalized on prime report showed that semantic priming was restricted to reported related primes. In Experiment 2, subjects made no overt response to the primes. Priming was conditionalized on recognition of the primes on a subsequent test. The pattern was the same as Experiment 1: There was priming only for recognized primes; recognition memory showed a pattern consistent with retroactive priming. Experiment 3 also conditionalized priming on recognition performance but used a prime-target interval of only 250 msec. Again, semantic priming was found only for recognized primes, and recognition memory revealed retroactive priming. Retroactive priming indicates an interdependency between prime and target processing that needs to be incorporated into models of semantic priming.     

Repetition proportion biases masked priming of lexical decisions

Although subjects have little or no awareness of masked primes, Bodner and Masson (2001) found that priming of lexical decisions is often enhanced when masked repetition primes occur on a high proportion of trials. We used baseline prime conditions to specify the locus of this repetition proportion (RP) effect. In Experiments 1A and 1B, a .8-RP group showed more priming than did a .2-RP group, and this RP effect was due to both (1) increased facilitation from repetition primes and (2) increased interference from unrelated primes. In Experiment 2, we used the baseline condition to show that subjects are sensitive to RP rather than to the proportion of unrelated primes. Direct comparisons of a given prime condition (repetition, unrelated) across RP conditions were more stable than were comparisons relative to the baseline condition. The increased costs and benefits of repetition priming when RP is higher implicate a context-sensitive mechanism that constrains accounts of masked priming.     

Beyond spreading activation: An influence of relatedness proportion on masked semantic priming

Semantic priming in the lexical decision task has been shown to increase when the proportion of related-prime trials is increased. This finding typically is taken as evidence for a conscious, strategic use of primes. Three experiments are reported in which masked semantic primes displayed for only 45 msec were tested in high- versus low-relatedness proportion conditions. Relatedness proportion was increased either by using a high proportion of semantically related primes or a large set of repetitionprimed filler trials. Semantic priming was consistently enhanced relative to a low-relatedness proportion condition. These relatedness proportion effects were not due to conscious, strategic use of primes: Exclusion of prime-aware subjects did not attenuate the effects, and better performance in a prime classification task was not associated with larger semantic priming effects. These results are interpreted within a retrospective account of semantic priming in which recruitment of a prime event is modulated by prime validity.     

Masked inhibitory priming in english: evidence for lexical inhibition

Predictions derived from the interactive activation (IA) model were tested in 3 experiments using the masked priming technique in the lexical decision task. Experiment 1 showed a strong effect of prime lexicality: Classifications of target words were facilitated by orthographically related nonword primes (relative to unrelated nonword primes) but were inhibited by orthographically related word primes (relative to unrelated word primes). Experiment 2 confirmed IA's prediction that inhibitory priming effects are greater when the prime and target share a neighbor. Experiment 3 showed a minimal effect of target word neighborhood size (N) on inhibitory priming but a trend toward greater inhibition when nonword foils were high-N than when they were low-N. Simulations of 3 different versions of the IA model showed that the best fit to the data is produced when lexical inhibition is selective and when masking leads to reset of letter activities.     

Does<Emphasis Type="SmallCaps">LGHT</Emphasis> prime<Emphasis Type="SmallCaps">DARK</Emphasis>? Masked associative priming with addition neighbors

We present two masked priming lexical decision experiments in which we examined whether a nonword prime word would activate associative/semantic information from its corresponding addition neighbor (e.g.,lght-DARK via the addition neighborlight), producing associative/semantic priming. The rationale was the following: If a nonword prime with a missing letter produced a semantic/associative priming effect, this would clearly indicate that this nonword was activating the lexical/semantic representations of its base word, thereby reinforcing the models of visual-word recognition in which the orthographic representations produced bylght (orligt) andlight are quite similar (e.g., SOLAR, SERIOL, open-bigram, and overlap models). The results showed that the magnitude of the masked associative priming effect with subset primes was remarkably similar to that of the priming effect with the corresponding word prime. Furthermore, the magnitude of the associative priming effect was similar when the deleted letter was a vowel and when the deleted letter was a consonant.     

Does the proportion of associatively related pairs modulate the associative priming effect at very brief stimulus-onset asynchronies?

A number of experiments have shown that the magnitude of the associative priming effect increases substantially when there is a high proportion of associatively related pairs in the list when the stimulus-onset asynchrony (SOA) between prime and target is long (more than 400 ms). In the present series of experiments we manipulated the proportion of associatively related pairs when the SOA was very brief (less than 200 ms). If processing of a target word is facilitated automatically by the prior presentation of a related prime, the occurrence of priming should be unaffected by the proportion of related pairs in the list. Experiment 1 showed a robust relatedness proportion effect obtained in a double lexical decision task. Experiments 2-4 used the masked priming technique at several very short SOAs (66, 116, and 166 ms) in lexical decision and naming. The results showed a reliable associative priming effect in the two tasks, which did not differ as a function of the proportion of related pairs. Finally, Experiment 5 used unmasked primes at an 83-ms SOA in which the primes remained in view after the target presentation. As in Experiments 2-4, the associative effect was not modulated by the proportion of associatively related pairs. The implications of these results are discussed.     

The facilitation of lexical decisions by a prime occurring after the target

Three experiments are reported that investigate priming effects in a lexical decision task. when the prime was presented after the target. In Experiment 1, the lexical decision target was presented for 50 msec, followed 80 msec later by the prime. No significant facilitation of responses was observed in the related prime condition. In Experiment 2, the target was presented for 30 msec, followed 35 msec later by the prime. Targets followed by related primes were responded to significantly faster than targets with unrelated primes. Experiment 3 replicated the result of Experiment 2. The data are interpreted as supporting parallel processing of the prime and target in semantic priming experiments. The theoretical implications of the “backward” priming effect are discussed.     

Subthreshold semantic priming: Not a context effect

Bernstein, Bissonnette, Vyas, and Barclay (1989) showed that masked primes can be recognized more often when followed by a related target than when followed by an unrelated target. They suggested that this effect may allow subjects in subthreshold semantic priming experiments to recognize the masked primes on the related semantic priming trials, even when the subjects cannot recognize masked primes presented alone. The present experiment found that when masked primes are presented at recognition rates comparable to those used in published studies of subthreshold masked priming, the forced choice recognition of masked primes followed by related targets was similar to that of masked primes presented alone (25% vs. 22%). In addition, the recognition of masked primes on the forced choice trials was highly predictive of masked prime recognition on the semantic priming trials.     

Repeated masked category primes interfere with related exemplars: new evidence for negative semantic priming

In 4 experiments, the authors found evidence for negatively signed masked semantic priming effects (with category names as primes and exemplars as targets) using a new technique of presenting the masked primes. By rapidly interchanging prime and mask during the stimulus onset asynchrony, they increased the total prime exposure to a level comparable with that of a typical visible prime condition without increasing the number of participants having an awareness of the prime. The negative effect was observed for only low-dominance exemplars and not for high-dominance exemplars. The authors found it using lexical decision (Experiments 1 and 2), lexical decision with a response-window procedure (Experiment 3), and the pronunciation task (Experiment 4). The results are discussed with regard to different theories on semantic priming.     

Masked repetition priming and proportion effects under cognitive load.

The authors used a cognitive load manipulation (rehearsing a string of digits during the trial) to test the automaticity of (a) masked repetition priming and (b) the masked repetition proportion (RP) effect (i.e., greater priming when the proportion of repetition-prime trials is higher) in the lexical decision task. The RP (.2 vs. .8) was varied across blocks. Masked priming was not reduced under load compared with a no-load group. Surprisingly, only the load group showed an RP effect in response latencies, although the no-load group showed an RP effect in the error rates. Our results show that masked priming is automatic, yet the influence of masked primes can nonetheless be adjusted at an unconscious level. Implications for accounts of masked priming are discussed.     

Retroactive semantic priming in a lexical decision task

The present study reports two experiments that required subjects to name target items preceded by a masked prime. Additionally, and subsequent to the naming task, subjects were required to indicate whether or not the prime was a word, along with a confidence rating of their lexical decision. Experiment 1 demonstrates that the processing of masked primes is facilitated by related targets when such targets are presented either 100 or 200 msec after the onset of the prime. Experiment 2 extends the finding of “retroactive” priming to a 1000=msec separation in prime-target presentation (SOA). The extent of retroactive priming is not dependent on SOA between prime and target, nor is it affected by the prime-mask SOA, which varied from 10 to 180 msec. Priming of targets was also independent of prime-target and prime-mask SOA, providing that primes had been classified as words. For word primes classified as non-words there was no semantic priming on target naming reaction time. Implications of these findings with respect to the nature of retroactive priming and the current controversy concerning subliminal priming effects were discussed.     

Semantic priming in the pronunciation task: The role of prospective prime-generated expectancies

In the relatedness proportion effect, semantic priming increases with an increase in the probability that a word prime will be followed by a semantically related word target. This effect has frequently been obtained in the lexical decision task but not in a pronunciation task. In the present experiment, relatedness proportion was manipulated in two pronunciation tasks, one with and one without nonword targets, using category names as primes. In both tasks, a relatedness proportion effect occurred for high-dominance category exemplars but not for low-dominance category exemplars. These results converge with recent lexical decision results in suggesting that semantic priming in pronunciation is affected by a prospective prime-generated expectancy that is modulated by the relatedness proportion.     

Being forward not backward: lexical limits to masked priming

This study investigated whether masked priming is mediated by existing memory representations by determining whether nonwords targets would show repetition priming. To avoid the potential confound that nonword repetition priming would be obscured by a familiarity response bias, the standard lexical decision and naming tasks were modified to make targets unfamiliar. Participants were required to read a target string from right to left (i.e., "ECAF" should be read as "FACE") and then make a response. To examine if priming was based on lexical representations, repetition primes consisted of words when read forwards or backwards (e.g., "face", "ecaf") and nonwords (e.g., "pame", "emap"). Forward and backward primes were used to test if task instruction affected prime encoding. The lexical decision and naming tasks showed the same pattern of results: priming only occurred for forward primes with word targets (e.g., "face-ECAF"). Additional experiments to test if response priming affected the LDT indicated that the lexical status of the prime per se did not affect target responses. These results showed that the encoding of masked primes was unaffected by the novel task instruction and support the view that masked priming is due to the automatic triggering of pre-established computational processes based on stored information.     

Reading “glasses” will prime “vision,” but reading a pair of “glasses” will not

In a lexical decision task with two primes and a target, the target was preceded 300 msec by the second prime (P2) which in turn was preceded by a brief forward and backward masked first prime (P1). When P1 and P2 were unrelated, reaction times were faster when the target was related to P2 (e.g.,wave SALT ... pepper) than when the target was unrelated to P2 (and P1—e.g.,wave LOAN ... pepper). However, this semantic priming effect was reduced to statistically nonsignificant levels when P1 and P2 were repetitions of the same word. That is, priming did not occur forsalt SALT ... pepper relative toloan LOAN ... pepper. This reduction in priming was observed whether P2 and the target were strongly or weakly related. These findings raise problems for current accounts of semantic priming.     

Evidence for word length coding during visual word recognition

In a masked priming procedure manipulating orthographic neighbourhood size, the priming word activates a number of word candidates of which the target could be one. Whether the target is one of the candidates or not determines how quickly it is recognised. However, the efficiency of lexical processing may be markedly less if all possible candidates are activated. One solution to this problem is if the visual system uses prime length information to reduce the number of candidates to a more manageable amount. Here, we investigated in two masked priming experiments whether prime length and orthographic information combine to facilitate target word recognition. In Experiment 1, we showed that the efficiency of visual word recognition is not influenced by the length of primes alone. However, when combined with orthographically related primes, word length coding is preserved. In Experiment 2, we investigated whether length priming affects recognition of short and long words differently. Results showed that only short words benefit from a same-length orthographically related prime, and that the priming effect does not generalise to longer words. These results suggest that the length of a word is not an essential feature in lexical processing, but that it can facilitate recognition by constraining the activation of orthographically related words.     

Does jugde activate COURT? Transposed-letter similarity effects in masked associative priming

Transposed-letter (TL) nonwords (e.g., jugde) can be easily misperceived as words, a fact that is somewhat inconsistent with the letter-position-coding schemes employed by most current models of visual word recognition. To examine this issue further, we conducted four masked semantic/associative priming experiments, using a lexical decision task. In Experiment 1, the related primes could be words, TL-internal nonwords, or replacement-letter (RL) nonwords (e.g., judge, jugde, or judpe, respectively; the target would be COURT). Relative to an unrelated condition, masked TL-internal primes produced a significant semantic/associative priming effect, an effect that was only slightly smaller than the priming effect for word primes. No effect, however, was observed for RL-nonword primes. In Experiment 2, the TL-nonword primes were created by switching the two final letters of the primes (e.g., judeg). The results again showed a semantic/associative priming effect for word primes, but not for TL-final nonword primes or for RL-nonword primes. Experiment 3 replicated the associative/semantic priming effect for TL-internal nonword primes, with, again, no effect for TL-final nonword primes. Finally, Experiment 4 again failed to yield a priming effect for TL-final nonword primes. The implications of these results for the choice of a letter-position-coding scheme in visual word recognition models are discussed.     

With great expectations, can two "wrongs" prime a "right"?

The proportion of related prime-target pairs (relatedness proportion, RP) and prime-target stimulus onset asynchrony (SOA) was varied to determine the involvement of strategic priming mechanisms in the reduction in semantic priming that occurs when a target follows an unmasked prime that itself receives immediate repetition priming from a masked prime. At 300-ms and 1,200-ms SOAs, (a) strategic semantic priming was operating, in that priming from a nonrepeated prime increased as RP increased from .25 to .75, and (b) for both RPs, prime repetition reduced semantic priming. At a 167-ms SOA, (a) priming from a nonrepeated prime was unaffected by RP, suggesting that strategic priming was not operating, and (b) for both RPs, prime repetition did not reduce semantic priming. Because prime repetition did not reduce priming at the 167-ms SOA (when only spreading activation should have been mediating semantic priming), the reduction in semantic priming produced by prime repetition is not evidence against spreading activation automaticity. Possible mechanisms through which prime repetition reduces semantic priming are discussed.     

Dissociated backward priming effects in lexical decision and pronunciation tasks

Backward priming was examined at 150- and 500-msec prime-target stimulus onset asynchronies (SOAs) using visually presented primes and targets in lexical decision and pronunciation tasks. Two kinds of backward relations were used: compound items for which targets and primes formed a word in the backward direction (e.g., prime: HOP; target: bell), and noncompound items for which targets and primes did not form a word but were associatively related in the backward but not the forward direction (e.g., prime: BABY; target: stork). Results showed that backward priming effects were equivalent for compounds and noncompounds. However, for lexical decisions, backward priming occurred at both SOAs, whereas for pronunciation, it occurred only at the 150-msec SOA. We discuss how this SOA-dissociated backward priming effect in lexical decision and pronunciation tasks poses a serious challenge for all theories of semantic priming.     

Masked priming of words and nonwords in a naming task: Further evidence for a nonlexical basis for priming

Masked priming effects in word identification tasks such as lexical decision and word naming have been attributed to a lexical mechanism whereby the masked prime opens a lexical entry corresponding to the target word. Two experiments are reported in which masked repetition priming effects of similar magnitude were obtained with word and nonword targets in a naming task. Masked orthographic priming was more stable for word than for nonword targets, although morphological primes produced no advantage beyond that achieved by matched orthographic primes. These results, taken together with the recent finding that repetition priming of nonwords can be obtained in the lexical decision task, support the view that masked priming of words and nonwords has a nonlexical component. We suggest that masked primes can enhance target identification by contributing to the construction of an orthographic or a phonological representation of the target, regardless of the target's lexical status.     

Parts outweigh the whole (word) in unconscious analysis of meaning

In unconscious semantic priming, an unidentifiable visually masked word (the prime) facilitates semantic classification of a following visible related word (the target). Three experiments reported here provide evidence that masked primes are analyzed mainly at the level of word parts, not whole-word meaning. In Experiment 1, masked nonword primes composed of subword fragments of earlier-viewed targets functioned as effective evaluative primes. (For example, after repeated classification of the targets angel and warm , the nonword anrm acted as an evaluatively positive masked prime.) Experiment 2 showed that this part-word processing was potent enough to oppose analysis at the whole-word level. Thus, smile functioned as an evaluatively negative (!) masked prime after repeated classification of smut and bile . Experiment 3 found no priming when masked word primes contained no parts of earlier targets. These results suggest that robust unconscious priming (a) is driven by analysis of part-word information and (b) requires previous classification of visible targets that contain the fragments later serving as primes. Contrary to a widely held view, analysis of subliminal primes appears not to function at the level of analysis of complete words.