Masking Interrupts Figure-Ground Signals in V1

Masking Interrupts Figure-Ground Signals in V1 Lamme Zipser Spekreijse

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.

Masking Interrupts Figure-Ground Signals in V1 2002 Journal of Cognitive Neuroscience 10.1162/089892902320474490 lamme2002-masking-interrupts-figure-ground-s lamme2002-masking-interrupts-figure-ground-s.txt, the project's staged copy 1e35358b13f5b4677fc39db67c17505d7de5ec78 open-access neuroscience

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 mechanism Quotation withheld because this record is not cleared for redistribution. backward masking of figure–ground segregation Quotation withheld because this record is not cleared for redistribution. electrophysiology electrophysiology Quotation withheld because this record is not cleared for redistribution. mechanisms The mask replaces the target's information at the low level by the time descending signals arrive, so the recurrent V1-extrastriate loop that generates figure-ground modulation is cut off while the feedforward sweep is left intact. Quotation withheld because this record is not cleared for redistribution. mechanisms Perception proceeds in stages: a feedforward sweep first analyses elementary features, then recurrent interactions between V1 and higher areas bind them into a coherent percept, and only when that recurrent activity spreads into executive areas can a conscious report be made. about this quote: THE QUOTE STOPS WHERE THE EXTRACTED TEXT IS INTERRUPTED: after its last word the text runs into the axis labels and caption of Figure 6 ('…Figure 6. Orientation selectivity during pattern masking…'), so the words that completed the sentence are not continuous with it and are not quoted. Quotation withheld because this record is not cleared for redistribution. mechanisms Figure-ground modulation is necessary but not sufficient for report, because an independent decision stage is interposed between the V1 modulation and the animal's behavioural response. Quotation withheld because this record is not cleared for redistribution. Orientation selectivity at the multiunit recording sites was weak, limiting the sensitivity of the feature-signal measure. what this limits: the feature-signal analysis at these sites: ORIENTATION SELECTIVITY AT THESE 'MULTIUNIT' SITES IS NOT VERY STRONG Quotation withheld because this record is not cleared for redistribution. The texture orientation comparison used only two fixed orientations, which further weakened the measured selectivity. what this limits: the texture-orientation comparison specifically: THE SELECTIVITY IS EVEN LESS PRONOUNCED, IN PART BECAUSE THERE WERE ONLY TWO ORIENTATIONS, FIXED AT 45 AND 135 DEGREES, TO COMPARE Quotation withheld because this record is not cleared for redistribution. The conclusion is one inferential step from the measurement, and the premise is stated. what this limits: the claim that masking interrupts recurrent processing: the authors reason from the premise that figure-ground modulations depend on feedback from extrastriate areas, and say their results suggest the interruption rather than show it Quotation withheld because this record is not cleared for redistribution. The effect is selective: one signal is fully suppressed and another is unaffected. what this limits: any summary of what masking does to V1 activity here: figure-ground signals are selectively and fully suppressed at the intervals producing invisibility, while the initial response transients carrying the scene's features are not affected Quotation withheld because this record is not cleared for redistribution. the stimulus sets presented RANDOMISED OR INTERLEAVED rather than in blocks eliminates: electrode drift biasing the comparison - if one condition is recorded before another, any slow change in the recording is confounded with the condition, and interleaving spreads it across both · outcome (stated by the tagger, no quotation): held Quotation withheld because this record is not cleared for redistribution. figure – ground signals figure – ground signals Quotation withheld because this record is not cleared for redistribution. initial response transients initial response transients Quotation withheld because this record is not cleared for redistribution. recurrent interactions recurrent interactions Quotation withheld because this record is not cleared for redistribution. extrastriate areas extrastriate areas Quotation withheld because this record is not cleared for redistribution. Feedforward inhibitory models Feedforward inhibitory models Quotation withheld because this record is not cleared for redistribution. backward masking backward masking Quotation withheld because this record is not cleared for redistribution. the masking effect - SELECTIVE, AND BOTH HALVES ARE IN ONE SENTENCE: figure-ground signals are fully suppressed at exactly the intervals that make the target invisible, while the initial response transients are not affected, so 'masking reduces responses in V1' names one of two opposite results Quotation withheld because this record is not cleared for redistribution. the conclusion - HELD GIVEN AN IMPORTED PREMISE: the measurement is a suppression in V1 and the conclusion is about recurrent interactions with higher areas, reached because figure-ground modulations are taken to depend on feedback from extrastriate areas, and offered as a suggestion Quotation withheld because this record is not cleared for redistribution. backward masking - TWO RIVAL MODEL FAMILIES, NAMED: feedforward inhibitory models have the second stimulus inhibiting the first before it reaches a conscious level, while another family has masking interfering with reentrant or recurrent processing; the same phenomenon supports two different accounts of where it happens Quotation withheld because this record is not cleared for redistribution. replicates → Lamme, Supèr, Landman, Roelfsema, & Spekreijse (2000) Quotation withheld because this record is not cleared for redistribution. replicates → Supèr et al. (2001) Quotation withheld because this record is not cleared for redistribution. backward masking —causes→ figure – ground signals backward masking figure – ground signals holds when: at the target-mask intervals that make the target psychophysically invisible · direction on the measure: eliminates · strength of the relation: 0.7 Quotation withheld because this record is not cleared for redistribution. backward masking —causes→ initial response transients backward masking initial response transients direction on the measure: no-change · strength of the relation: 0.7 Quotation withheld because this record is not cleared for redistribution. t-test t-test Quotation withheld because this record is not cleared for redistribution. t-test t-test Quotation withheld because this record is not cleared for redistribution. p-value p-value Quotation withheld because this record is not cleared for redistribution. standard deviation standard deviation Quotation withheld because this record is not cleared for redistribution. standard error of the mean standard error of the mean Quotation withheld because this record is not cleared for redistribution. mean squared error mean squared error Quotation withheld because this record is not cleared for redistribution. median median Quotation withheld because this record is not cleared for redistribution. normalization normalization Quotation withheld because this record is not cleared for redistribution. Backward masking abolishes the figure-ground signal in V1 exactly at those target-mask intervals at which the monkey cannot see the figure. Quotation withheld because this record is not cleared for redistribution. The early transient V1 response that codes the local features of the scene survives the mask untouched. Quotation withheld because this record is not cleared for redistribution. With a texture pattern mask at short SOAs the monkeys performed at chance in locating the figure, confirming that the masks made the figure invisible. Quotation withheld because this record is not cleared for redistribution. Without a mask, figure-ground signals were still present at very short SOAs. Quotation withheld because this record is not cleared for redistribution. Orientation selectivity for the texture elements was essentially unaffected by backward masking at every SOA but the shortest. Quotation withheld because this record is not cleared for redistribution. Orientation information about the line segments remains available in V1 even at the SOAs at which the animals see nothing. Quotation withheld because this record is not cleared for redistribution. Figure-ground modulation tracks the psychometric curve far better (error 0.033) than orientation selectivity does (error 0.138). Quotation withheld because this record is not cleared for redistribution. Trial by trial, the amount of figure-ground modulation did not predict whether the monkey reported seeing the figure. Quotation withheld because this record is not cleared for redistribution. Detection of the figure occurs only when the stimulus also drives recurrent processing between V1 and higher areas, expressed as figure-ground modulation of V1 activity. Quotation withheld because this record is not cleared for redistribution. backward masking backward masking Quotation withheld because this record is not cleared for redistribution. recurrent processing recurrent processing Quotation withheld because this record is not cleared for redistribution. Figure – ground modulation Figure – ground modulation about this quote: THE QUOTE STOPS WHERE THE EXTRACTED TEXT IS INTERRUPTED: after its last word the text runs into the page's author running head and page number and then the labels of Figure 2 ('Lamme, Zipser, and Spekreijse 1045 Trial Figure-Ground…'), so the words that completed the sentence are not continuous with it and are not quoted. Quotation withheld because this record is not cleared for redistribution. feedforward inhibitory model feedforward inhibitory model Quotation withheld because this record is not cleared for redistribution. figure position figure position Quotation withheld because this record is not cleared for redistribution. percentage correct responses percentage correct responses Quotation withheld because this record is not cleared for redistribution. orientation selectivity orientation selectivity Quotation withheld because this record is not cleared for redistribution. mask type mask type Quotation withheld because this record is not cleared for redistribution. 0.033 — mean squared difference between the figure-ground neurometric curve and the behavioural psychometric curve Quotation withheld because this record is not cleared for redistribution. 0.138 — mean squared error between the orientation-selectivity neurometric curve and the same psychometric curve, i.e. the much poorer feedforward fit Quotation withheld because this record is not cleared for redistribution. p < .025 — significance threshold at which figure-ground modulation was present only at SOAs of 5-8 frames under pattern masking Quotation withheld because this record is not cleared for redistribution. 25 of these 32 electrodes — number of implanted V1 recording channels contributing data to this study Quotation withheld because this record is not cleared for redistribution. 700 (total 11,000) — trials collected per SOA/mask condition, and in total, for the neural analysis Quotation withheld because this record is not cleared for redistribution. 1.94 — median orientation-selectivity ratio of the multiunit recording sites Quotation withheld because this record is not cleared for redistribution. 14 – 110 msec — range of stimulus-onset asynchronies (1 to 8 monitor frames) between figure-ground display and mask Quotation withheld because this record is not cleared for redistribution. 100 msec — latency at which V1 responses start to distinguish figure from background texture Quotation withheld because this record is not cleared for redistribution. 0.18 — maximum figure-ground modulation, equated to 100% correct when building the neurometric curve 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/10.1162/089892902320474490.