| phenomenon | semantic priming in lexical decision— “The commonly obtained finding in such experiments (e.g., Meyer & Schvaneveldt, 1971; Meyer, Schvaneveldt, & Ruddy, Note 1) is that subjects are quicker to respond that a target letter string (e.g., NURSE) is a word when the immediately prior target was a semantically related word (i.e., DOCTOR) than when it was a semantically… |
| claim-type | mechanism— “These results were regarded as supporting the theory of Posner and Snyder that postulates two distinct components of attention: a fast automatic inhibitionless spreading-activation process and a slow limited-capacity consciousattention mechanism” |
| method — evidence | — “Prior to each visually presented target letter string in a speeded word-nonword classification task, either BIRD, BODY, BUILDING, or xxx appeared as a priming event” |
| builds-on | Posner and Snyder (1975a) two-factor theory of attention— “The word-target data confirmed all five of the major predictions derived from Posner and Snyder's (1975a) two-factor theory of attention and therefore offer striking support for that theory.” | claim_relations[0] |
| extends | Neely (1976)— “The purpose of the present research was to provide a more rigorous test of the basic assumptions underlying the Posner-Snyder (197Sa) theory by elaborating on the Neely (1976) paradigm so as to permit a further clarification of the roles of limited-capacity attention and automatic spreading activation in the lexical decision task” | claim_relations[1] |
| extends | Rosch (1975)— “This finding extends the generality of a result obtained by Rosch (1975).” | claim_relations[2] |
| sense | THE PAPER'S MOVE IS TO MAKE THE CONFOUND INTO THE DESIGN. Earlier work could not say how much facilitation came from attention and how much from automatic spreading, because relatedness and expectancy travel together: a prime that is related to the target is also a prime that makes it expected. Neely breaks them apart - `the presence or absence of automatic spreading activation was controlled by whether the prime and the target were semantically related`, while conscious attention `was controlled by variations in the PROBABILITY that the priming cue itself would appear`. Two knobs where the field had one— “First, the presence or absence of automatic spreading activation was controlled by whether the prime and the target were semantically related or semantically unrelated” | senses[0] |
| sense | AND THE CELL THAT PAYS FOR IT IS BODY-heart: A SIGN REVERSAL PRODUCED BY NOTHING BUT DELAY. The prime is semantically related to the target, but the subject expects a different category - so automatic activation says facilitate and conscious attention says inhibit, in the same trial. `there was an inhibition effect at the 2,000 msec SOA, which decreased as the SOA decreased until, at the 250-msec SOA, it became a facilitation effect.` Not a difference in size between conditions - the same condition changing sign as the slow mechanism is given less time to act. That is as close to a crossing double dissociation as a priming experiment gets— “On BODY-heart type trials, there was an inhibition effect at the 2,000 msec SOA, which decreased as the SOA decreased until, at the 250-msec SOA, it became a facilitation effect” | senses[1] |
| sense | THE ASYMMETRY THAT MAKES THE INFERENCE WORK, STATED AS A PROPERTY RATHER THAN A RESULT: `Automatic spreading activation produces only facilitatory effects and no inhibitory effects.` Both mechanisms facilitate; only one inhibits. So facilitation is evidence of nothing in particular and inhibition is evidence of attention - which is why the whole design is built to produce inhibition and why the 250-msec SOA, where inhibition vanishes, is the control rather than a null result— “Automatic spreading activation produces only facilitatory effects and no inhibitory effects” | senses[2] |
| sense | AND IT IS THE FIFTH RECORD RUNNING WHERE THE INSTRUMENT IS PART OF THE PHENOMENON - BUT THE FIRST WHERE THE AUTHOR KNEW. Finkbeiner2011: task difficulty decides what counts as subliminal. erb2019: associative overlap decides whether a CSE is control. eckstein2007: which awareness test decides whether the prime was seen. lukatela2007: within- vs between-subjects decides whether the lexicons look separate. Here SOA decides WHICH OF TWO MECHANISMS you are measuring - and Neely makes that the independent variable instead of discovering it afterwards. The corpus now holds the 1977 paper that turned the artefact into the design, and the 2007 paper that found the same class of artefact still undetected in its own neighbouring literature.— “The stimulus onset asynchrony (SOA) between the prime and the target letter string varied between 250 and 2,000 msec” | senses[3] |
| sense | AND THE MECHANISM CLAIM IS SHARPER THAN `ATTENTION SLOWS THINGS DOWN`: `the conscious-attention mechanism does not inhibit the buildup of activation in unattended logogens but rather inhibits the readout of information from unattended logogens.` The activation still accumulates; what attention withholds is access to it. That is a falsifiable division of labour - it predicts an unattended item can be primed without being reportable, which is the exact question Finkbeiner2011, eckstein2007 and hesselmann2016 are still arguing about thirty to forty years later, and none of the three cites this sentence for it— “According to their theory, the conscious-attention mechanism does not inhibit the buildup of activation in unattended logogens but rather inhibits the readout of information from unattended logogens” | senses[4] |
| control | the stimulus onset asynchrony held constant at 2,000 msec for every trial in the 2,000-alone group — held — rules out the SOA varying within the group, which would confound the expectancy manipulation with the time available to form the expectancy— “In the 2,000-alone group, the SOA was held constant at 2,000 msec for all trials.” | controls[0] |
| audit | conscious attention —causes→ expectancy — Audit 2026-09-28: DOUBTFUL - read against its span and kept, but the span says the expectancy is generated via conscious attention; the sign 'increases' is not stated.— “According to this semantic-matching hypothesis, the subject has a tendency to respond on the basis of the similarity between the semantic features of the expectancy generated, via conscious attention, to the word prime and the semantic features activated by the target letter string. If the semantic features” | concept_edges[0] |
| effect direction (stated by the tagger, no quotation) | conscious attention —causes→ expectancy — increases | concept_edges[0] |
| note | conscious attention —causes→ expectancy — THE PHRASE THAT CARRIES THE EDGE IS A FOUR-WORD ASIDE: 'the expectancy generated…'. Not a sentence about how expectancies arise - a subordinate clause inside a sentence about semantic matching. An earlier note of ours says the sentence carrying the mechanism usually cannot carry the edge; here the edge is carried by a parenthetical inside a sentence that is about something else entirely. Weight 0.9: the relation is asserted rather than argued, and it is asserted while the author is busy with a different claim. The same sentence cites Collins & Loftus 1975, which this corpus merged four hours ago - the spreading-activation paper whose mechanism Neely is here separating from attention.— “According to this semantic-matching hypothesis, the subject has a tendency to respond on the basis of the similarity between the semantic features of the expectancy generated, via conscious attention, to the word prime and the semantic features activated by the target letter string. If the semantic features” | concept_edges[0] |
| weight | conscious attention —causes→ expectancy — 0.9— “According to this semantic-matching hypothesis, the subject has a tendency to respond on the basis of the similarity between the semantic features of the expectancy generated, via conscious attention, to the word prime and the semantic features activated by the target letter string. If the semantic features” | concept_edges[0] |
| note | facilitation —contrasted-with→ inhibition — THE DISSOCIATION IN ONE SENTENCE, AND IT IS THE PAPER'S WHOLE ARGUMENT. 'facilitation occurred automatically…'. Regardless / only - the two quantifiers are what turn a pair of effects into a pair of mechanisms. Weight 0.95: this is a result sentence, it names both endpoints, and the contrast it draws is the one the whole design was built to make. Rare on two counts - says the sentence carrying the mechanism usually cannot carry the edge, and it carried both here; and it is the first edge today whose evidence needed no trimming at all.— “Thus, as predicted, facilitation occurred automatically, regardless of whether the subject was consciously using the priming cue to direct his or her attention to the logogen for a target element, whereas inhibition occurred only when the subject was misdirecting attention” | concept_edges[1] |
| weight | facilitation —contrasted-with→ inhibition — 0.95— “Thus, as predicted, facilitation occurred automatically, regardless of whether the subject was consciously using the priming cue to direct his or her attention to the logogen for a target element, whereas inhibition occurred only when the subject was misdirecting attention” | concept_edges[1] |
| audit | automatic spreading activation —causes→ facilitation — Audit 2026-09-28: DOUBTFUL - read against its full span and kept, but the span says the effect CAN BE ACCOUNTED FOR by assuming this; hedged.— “Such an effect can be accounted for by assuming that the facilitation effect produced by automatic spreading activation is slightly greater than the facilitation effect produced by limited-capacity attention” | concept_edges[2] |
| effect direction (stated by the tagger, no quotation) | automatic spreading activation —causes→ facilitation — increases | concept_edges[2] |
| note | automatic spreading activation —causes→ facilitation — BOTH MECHANISMS FACILITATE, AND THE QUOTE IS COMPARING THEIR SIZES. That is worth keeping rather than flattening: the edge is real - automatic spreading activation produces facilitation - but the sentence exists to say that its facilitation is 'slightly greater' than the attentional kind. Weight 0.9 rather than 0.95 because the claim is made inside an 'assuming that' - Neely is deriving what would have to be true for an observed pattern, not reporting a measurement. The hedge is in the quote and I have not cut it out.— “Such an effect can be accounted for by assuming that the facilitation effect produced by automatic spreading activation is slightly greater than the facilitation effect produced by limited-capacity attention” | concept_edges[2] |
| weight | automatic spreading activation —causes→ facilitation — 0.9— “Such an effect can be accounted for by assuming that the facilitation effect produced by automatic spreading activation is slightly greater than the facilitation effect produced by limited-capacity attention” | concept_edges[2] |
| outcome (stated by the tagger, no quotation) | control — held | controls[0] |
| replication | untested |
| limitation | Whether the inhibition in the Unexpected conditions is due solely to a semantic-matching strategy remains unresolved because of contrary evidence at the 700-msec SOA.— “Unfortunately, the issue of whether the inhibition effect obtained for the Unexpected word-target conditions can be solely attributed to the operation of a semantic-matching strategy remains open because a substantial inhibition effect of 56 msec was obtained in Condition Shift-Unexpected-Unrelated at the 700-msec SOA, at which there was no… | limitations[0] |
| limitation | The claimed asymmetry in buildup of conscious-attention facilitation versus inhibition may be only a failure to detect an effect.— “Of course, these results may merely reflect a failure to detect a facilitation effect for BODY-door trials at the 400-nisec SOA, but if future research should further substantiate this asymmetry in the buildup of consciousattention facilitation and conscious-attention inhibition, it would be theoretically quite important” | limitations[1] |
| limitation | The data cannot adequately answer in what sense activation is 'spreading' if high- and low-dominance exemplars are facilitated equally.— “Given the available data, it is difficult to answer this question adequately” | limitations[2] |