Processing of natural images is feedforward: a simple behavioral test

Processing of natural images is feedforward: a simple behavioral test Schmidt Schmidt

Tag layer of the DHSS Meta-Tagging Project. The article text is reproduced from the staged copy named in sourceDesc and is not altered; the tags are held separately in standOff so that overlapping and nested spans are preserved.

Processing of natural images is feedforward: a simple behavioral test 2009 Attention, Perception & Psychophysics 10.3758/app.71.3.594 Schmidt2009-processing-of-natural-images-is-feedforward Schmidt2009-processing-of-natural-images-is-feedforward.txt, the project's staged copy dee846a8e2a4de4c054cf3ddde910f2c28cd545a not-redistributable cognitive-psychology

the paper's own words

the tag layer; a sentence written by the tagger about the paper, always anchored to a quotation but not itself a quotation

0 spans located by searching the stored text for the quote's exact words, allowing any whitespace between them existence Quotation withheld because this record is not cleared for redistribution. visuomotor response priming with natural images Quotation withheld because this record is not cleared for redistribution. behavioural-experiment behavioural-experiment Quotation withheld because this record is not cleared for redistribution. mechanisms The prime's feedforward wave arrives at motor areas first and drives the movement alone until the target's wave arrives and takes over; a longer SOA gives the prime a longer unopposed drive, which is why priming grows with SOA. Quotation withheld because this record is not cleared for redistribution. mechanisms When no overt detour appears, opposing prime- and target-driven response tendencies conflict covertly and must be resolved before the movement can start, which shows up as a delayed onset instead. Quotation withheld because this record is not cleared for redistribution. The paper does not claim that the rapid classification is driven by image semantics specifically. Quotation withheld because this record is not cleared for redistribution. In the slow large-small task the independence criterion could not be clearly evaluated. Quotation withheld because this record is not cleared for redistribution. In Experiment 1 the raw trajectories did not show the initiation criterion; the conclusion rests on the derived spatial priming functions. Quotation withheld because this record is not cleared for redistribution. Meeting the rapid-chase criteria shows behavioural equivalence to a feedforward system, not that every processing substage is feedforward. Quotation withheld because this record is not cleared for redistribution. eight sessions per participant, four in each task, with task order counterbalanced eliminates: that a difference between the two tasks is an order effect · outcome (stated by the tagger, no quotation): held Quotation withheld because this record is not cleared for redistribution. rapid-chase criteria rapid-chase criteria Quotation withheld because this record is not cleared for redistribution. feedforward feedforward Quotation withheld because this record is not cleared for redistribution. independence criterion independence criterion Quotation withheld because this record is not cleared for redistribution. priming effects priming effects Quotation withheld because this record is not cleared for redistribution. THREE NAMED CRITERIA, EACH WITH ITS JOB STATED - THE CLEAREST `a tool must be checkable` IN THIS CORPUS: `(1) Prime rather than target signals should determine the onset and initial direction of the response (initiation criterion); (2) target signals should be able to influence the response before it is completed (takeover criterion); and, crucially, (3) response kinematics should initially depend exclusively on prime characteristics and be independent of all target characteristics (independence criterion).` Numbered, named, and separately falsifiable - a dataset can meet 1 and 2 and fail 3, which is exactly what happens in one of this paper's four tasks. Compare jitendra2016, which grades a literature against someone else's published checklist: same discipline, applied to a mechanism instead of a field Quotation withheld because this record is not cleared for redistribution. AND THE CLAIM IS EQUIVALENCE, NOT IDENTITY, STATED BY THE AUTHORS IN THEIR OWN CONCLUSION: `our data indicate a rapid-chase system ... that is behaviorally equivalent to a feedforward system with high temporal resolution.` They say elsewhere that the criteria `do not guarantee that a stimulus-response system is strictly feedforward in all its substages`. A behavioural test that names the limit of what behaviour can establish. Most records in this corpus infer a mechanism from a behavioural measure; this one declines to, and says which claim the method does support instead Quotation withheld because this record is not cleared for redistribution. THE PAPER CONTAINS ITS OWN BOUNDARY CONDITION - ONE OF FOUR TASKS FAILS: `Only in the large-small task, with its tardy response dynamics and small priming effects, does it become difficult to evaluate whether the independence criterion is met.` And the diagnosis is offered rather than withheld: `It rather seems that the large-small classification cannot be performed in a single feedforward sweep and requires more extensive, recurrent processing.` Same images, same participants, same apparatus - only the classification rule changed, and the account stops applying. A scope limit measured inside the paper rather than conceded to a critic later Quotation withheld because this record is not cleared for redistribution. AND THE DOSE-RESPONSE COMES WITH ITS MECHANISM ATTACHED, WHICH IS RARE: `Priming effects increase with prime-target SOA because the prime has more time to drive the response on its own when the target is further delayed.` The `because` is not decoration - it is the rapid-chase model making the prediction, and the effect size is a duration rather than a strength. dijksterhuis1998 doses by prime DURATION and gets a bigger effect; this doses by the GAP and gets a longer unopposed run - two different quantities that both look like `more priming` in a results table Quotation withheld because this record is not cleared for redistribution. extends → the authors' own rapid-chase theory of response priming (Schmidt et al., 2006; Vath & Schmidt, 2007) Quotation withheld because this record is not cleared for redistribution. builds-on → rapid natural-scene categorisation studies (Bacon-Macé et al., 2007; Kirchner & Thorpe, 2006; Thorpe et al., 1996; VanRullen & Koch, 2003; VanRullen & Thorpe, 2001, 2002) Quotation withheld because this record is not cleared for redistribution. contradicts → the claim that natural image classification is "ultra-rapid" (Kirchner & Thorpe, 2006; VanRullen & Thorpe, 2001) Quotation withheld because this record is not cleared for redistribution. feedforward —operationalised-as→ rapid-chase criteria feedforward rapid-chase criteria strength of the relation: 0.95 · note: THE WHOLE PAPER IN ONE EDGE, AND 'would show in the time course of the response' IS THE OPERATIONALISATION. A claim about neuronal signal flow is turned into three tests on a pointing trajectory. Weight 0.95: the criteria are named, numbered and separately checkable, and the authors state in the same passage that meeting them establishes BEHAVIOURAL EQUIVALENCE and not strict feedforward - so the edge is exactly as strong as the paper says it is, which is why it earns 0.95 rather than being hedged down. Quotation withheld because this record is not cleared for redistribution. independence criterion —part-of→ rapid-chase criteria independence criterion rapid-chase criteria strength of the relation: 0.95 · note: THE THIRD OF THREE, and the quote carries all three in order with their names. Recorded separately because it is the one that FAILS in the large-small task - a corpus that holds only 'rapid-chase criteria' as one node cannot express that a system met two of them. Weight 0.95: the membership is explicit in the enumeration. Quotation withheld because this record is not cleared for redistribution. F-test / ANOVA F-test / ANOVA Quotation withheld because this record is not cleared for redistribution. jackknife jackknife Quotation withheld because this record is not cleared for redistribution. arcsine transformation arcsine transformation Quotation withheld because this record is not cleared for redistribution. data smoothing filter data smoothing filter Quotation withheld because this record is not cleared for redistribution. standard error standard error Quotation withheld because this record is not cleared for redistribution. In the animal-versus-object task, the time course of spatial priming matched the pattern the rapid-chase model predicts. Quotation withheld because this record is not cleared for redistribution. Spatial priming curves for every prime-target SOA first followed one identical trajectory and branched off only later, in order of increasing SOA, in line with the independence criterion. Quotation withheld because this record is not cleared for redistribution. The large-versus-small size-judgement task was the one case where the independence criterion could not be assessed, because its responses were slow and its priming effects small. Quotation withheld because this record is not cleared for redistribution. The authors attribute the slowness of the size-judgement task to its needing recurrent processing rather than a single feedforward sweep. Quotation withheld because this record is not cleared for redistribution. There was no indication that natural-image categorisation is extraordinarily fast, or 'ultra-rapid'. Quotation withheld because this record is not cleared for redistribution. When prime and target categories conflicted, the pointing hand physically veered into the quadrant indicated by the prime before reversing toward the correct target. Quotation withheld because this record is not cleared for redistribution. Pointing errors occurred almost only on inconsistent trials and grew with prime-target SOA, showing prime influence reaches the motor level. Quotation withheld because this record is not cleared for redistribution. The authors conclude that visuomotor priming reads out the first pass of information through the visuomotor system, before recurrent processing contributes. Quotation withheld because this record is not cleared for redistribution. feedforward processing feedforward processing Quotation withheld because this record is not cleared for redistribution. rapid chase rapid chase Quotation withheld because this record is not cleared for redistribution. rapid-chase criteria rapid-chase criteria Quotation withheld because this record is not cleared for redistribution. independence criterion independence criterion Quotation withheld because this record is not cleared for redistribution. spatial priming function spatial priming function Quotation withheld because this record is not cleared for redistribution. prime–target SOA prime–target SOA Quotation withheld because this record is not cleared for redistribution. pointing trajectories pointing trajectories Quotation withheld because this record is not cleared for redistribution. rates of response errors rates of response errors Quotation withheld because this record is not cleared for redistribution. 380 msec — time after prime onset at which the hand began moving on consistent trials in the large-small task (versus roughly 300 msec in the animal-object task) about this quote: THE QUOTE CONTINUES A SENTENCE THAT THIS COPY CUTS IN TWO - a running head and the caption of Figure 3 stand between it and the words it follows, which read 'roughly 300 msec after prime onset in the'. Quotation withheld because this record is not cleared for redistribution. 280 msec / 220 msec — movement onset times after prime onset in the toy-animal-object and ellipse-rectangle tasks of Experiment 2 Quotation withheld because this record is not cleared for redistribution. 160 msec — lag of priming-effect onset in the large-small task relative to the ellipse-rectangle task Quotation withheld because this record is not cleared for redistribution. 33, 50, 67, 83, or 100 msec — the five prime-target stimulus onset asynchronies used, with primes shown for 33 msec Quotation withheld because this record is not cleared for redistribution. Twelve — total participants, six in each of the two experiments Quotation withheld because this record is not cleared for redistribution. 1.14% and 1.73% — trials discarded by the exclusion rules in Experiments 1 and 2 Quotation withheld because this record is not cleared for redistribution. about 80 msec — how much earlier spatial priming onset occurred in the animal-object task than in the large-small task Quotation withheld because this record is not cleared for redistribution.

The full text of this article is not redistributable, so it is not included here. The evidence sentences are withheld with it; the tag layer and its grounding status remain. The article itself is at https://doi.org/10.3758/app.71.3.594.