phenomenon · claim-type · cross-domain-transfer · mechanism · claim-relation · replication-status · limitation — grounded in the full text| phenomenon | cortical multisensory processing— “While cortical multisensory processing has been extensively investigated in terms of psychophysics and mesoscale neural correlates, an in depth understanding of the underlying circuit-level mechanisms is lacking.” | |
| claim-type | mechanism— “We conclude that distinct multisensory processes share overlapping anatomical substrates, are implemented by diverse neuronal micro-circuitries that operate in parallel, and are flexibly recruited based on factors such as stimulus features and behavioral constraints.” | |
| method — evidence | — “In this review, we expand the framework on the circuit-level mechanisms of cortical multisensory processing by highlighting that multisensory processing is a family of functions – rather than a single operation – which involves not only the integration but also the segregation of modalities.” | |
| mechanism | sensory noise is independent across sense organs but shared within one modality, so pooling across modalities cancels more noise and yields a bigger behavioral gain than pooling within a modality— “A possible explanation is that sensory noise is largely non-overlapping between different sensory organs whereas it is mostly shared within a specific sensory modality.” | mechanisms[0] |
| mechanism | sound enhances vision by disinhibition: A1 layer-5 axons drive layer-1 interneurons in V1, which inhibit the layer 2-3 interneurons that normally suppress pyramidal cells— “Excitatory influences were mediated by projections from layer 5 in A1 to layer 1 interneurons in V1; these interneurons would then further inhibit layer 2–3 interneurons impinging on pyramidal cells, which would thus be disinhibited, resulting in enhanced visual responses” | mechanisms[1] |
| mechanism | a cross-modal stimulus resets ongoing oscillations, imposing a cycle of high and low excitability so that later same-modality inputs are amplified or depressed according to oscillation phase— “In this way, the stimulus instantiated a rhythmic fluctuation of the excitability of the auditory cortical circuit (mediated via neurons in supragranular layers), by which responses to auditory stimuli occurring at the peaks of the oscillation were enhanced and responses occurring at the troughs were depressed.” | mechanisms[2] |
| mechanism | prefrontal input to the thalamic reticular formation sets thalamic sensory gain, gating how much of each modality reaches primary cortex and thereby implementing sensory selection— “The reticular thalamic formation acts as a relay station which – based on the inputs received from the medial prefrontal cortex – is able to selectively control thalamic sensory gain and therefore gate the level of stimulation that reaches primary cortices (Wimmer et al., 2015)” | mechanisms[3] |
| mechanism | selecting neurons at the low tail of a normal response distribution guarantees they drift upward on remeasurement, manufacturing an apparent inverse-effectiveness effect out of nothing— “Therefore, even without any quantitative difference between condition A and B, neurons that have been selected to be on the low end of the normal distribution in condition A will ‘regress to the mean’ in condition B resulting in a seemingly higher response and vice versa.” | mechanisms[4] |
| mechanism | a loud auditory distractor overshadows a concurrent visual cue, capturing processing resources and cutting the behavioral response to the visual stimulus— “This indicates that the loud auditory noise overshadowed the visual stimulus, resulting in a reduced behavioral output in response to the visual cue.” | mechanisms[5] |
| builds-on | Bizley et al. (2016)— “This view is implicit to the definition of multisensory or cross-modal integration as any process in which information across sensory modalities is combined to make a perceptual inference (Bizley et al., 2016).” | claim_relations[0] |
| contradicts | van Atteveld et al. (2014)— “However, while some forms of MP (e.g. cue integration) might be fully explained by the proposed two mechanisms, we believe that a complex, diverse phenomenon such as MP can only be explained by considering a larger set of computational rules.” | claim_relations[1] |
| sense | multisensory processing - A FAMILY OF FUNCTIONS, NOT ONE OPERATION: the review's central move is to say the term covers not only integration but also SEGREGATION of modalities, so a word the field uses for combining also names the deciding-not-to-combine, and the two have opposite outcomes— “by highlighting that multisensory processing is a family of functions – rather than a single operation – which involves not only the integration but also the segregation of modalities” | senses[0] |
| sense | the MLE reliability claim - TRUE BY CONSTRUCTION, AND LABELLED SO: the combined estimate's reliability is the sum of the unisensory reliabilities and is therefore by definition more reliable; the empirical claim that the framework applies is separate, and only the second can fail— “According to the MLE rule, the reliability of the combined multisensory estimate is the sum of each unisensory reliability and is therefore by definition more reliable” | senses[1] |
| sense | the definition in use - CHOSEN FOR ITS BREADTH: any process in which information across sensory modalities is combined to make a perceptual inference, attributed to Bizley et al., and stated to encompass functions that both precede and incorporate cue integration— “This view is implicit to the definition of multisensory or cross-modal integration as any process in which information across sensory modalities is combined to make a perceptual inference (Bizley et al., 2016). This definition encompasses functions which both precede and incorporate cue integration” | senses[2] |
| control | the observed correlation was tested against a NULL DISTRIBUTION built for the purpose — held — rules out REGRESSION TO THE MEAN - that a correlation between unisensory responsiveness and multisensory modulation appears simply because extreme values move toward the mean— “First, to control for the possibility of regression to the mean, the observed correlation between unisensory responsiveness and multisensory response modulation should be tested against a null-distribution which is obtained by a shuffling procedure.” | controls[0] |
| control | neurons with the lowest and highest responses were removed and inverse effectiveness re-checked on what remained — held — rules out that inverse effectiveness is a CEILING AND FLOOR artefact - neurons operating at the limits of their dynamic range cannot show further modulation— “Second, to control for the potential problem of neurons operating at the limits of their dynamic range, it should be assured that inverse effectiveness persists after removal of neurons with the lowest and highest firing rates.” | controls[1] |
| edge | integration —part-of→ multisensory processing— “by highlighting that multisensory processing is a family of functions – rather than a single operation – which involves not only the integration but also the segregation of modalities” | concept_edges[0] |
| edge | segregation —part-of→ multisensory processing— “by highlighting that multisensory processing is a family of functions – rather than a single operation – which involves not only the integration but also the segregation of modalities” | concept_edges[1] |
| outcome (stated by the tagger, no quotation) | control — held | controls[0] |
| outcome (stated by the tagger, no quotation) | control — held | controls[1] |
| for which use | limitation — the circuit inventory the review assembles: THE PICTURE OF AVAILABLE MICRO-CIRCUITRIES STARTS EMERGING FROM RECENT RESEARCH BUT IS FAR FROM BEING UNAMBIGUOUS AND COMPLETE— “The picture of available micro-circuitries starts emerging from recent research but is far from being unambiguous and complete.” | limitations[0] |
| for which use | limitation — the level of generality of every mechanism reviewed: WHETHER SPECIFIC MECHANISMS PERTAIN TO ASSOCIATION CORTEX, SENSORY CORTEX, OR REPRESENT MORE GENERAL CODING SCHEMES AWAITS FURTHER INVESTIGATION— “Whether specific mechanisms pertain to association cortex, sensory cortex, or represent more general coding schemes awaits further investigation.” | limitations[1] |
| for which use | limitation — the hierarchical-Bayesian framing: WHETHER SUCH SPECIFIC CIRCUIT IMPLEMENTATIONS ARE COMPATIBLE WITH A GENERAL MECHANISM FOR MP, AND CAN BE MODELED IN A HIERARCHICAL BAYESIAN FASHION ... REMAINS TO BE ADDRESSED— “Whether such specific circuit implementations are compatible with a general mechanism for MP, and can be modeled in a hierarchical Bayesian fashion (Box 1) remains to be addressed.” | limitations[2] |
| for which use | limitation — any claim that MLE integration was confirmed to improve reliability: under the MLE rule the combined estimate's reliability is the sum of the unisensory reliabilities and is therefore by definition more reliable— “According to the MLE rule, the reliability of the combined multisensory estimate is the sum of each unisensory reliability and is therefore by definition more reliable” | limitations[3] |
| for which use | limitation — any count or comparison of multisensory findings under this review: the definition used is any process in which information across sensory modalities is combined to make a perceptual inference, which encompasses functions that both precede and incorporate cue integration— “This view is implicit to the definition of multisensory or cross-modal integration as any process in which information across sensory modalities is combined to make a perceptual inference (Bizley et al., 2016). This definition encompasses functions which both precede and incorporate cue integration” | limitations[4] |
| for which use | limitation — the evidence base of this review: studies in cats and non-human primates are included if no rodent data are available or if the results have been seminal to the field— “We will refer to studies with cats and non-human primates if no rodent data are available in the literature, or if results have been seminal to the field” | limitations[5] |
| replication | untested | |
| limitation | The inventory of micro-circuits supporting multisensory processing is still incomplete and ambiguous.— “The picture of available micro-circuitries starts emerging from recent research but is far from being unambiguous and complete.” | limitations[0] |
| limitation | It is not yet known whether the mechanisms reviewed are specific to association or sensory cortex or are general coding schemes.— “Whether specific mechanisms pertain to association cortex, sensory cortex, or represent more general coding schemes awaits further investigation.” | limitations[1] |
| limitation | Whether the specific A1-V1 circuit implementations fit a general mechanism or a hierarchical Bayesian model is unresolved.— “Whether such specific circuit implementations are compatible with a general mechanism for MP, and can be modeled in a hierarchical Bayesian fashion (Box 1) remains to be addressed.” | limitations[2] |
| limitation | The model's reliability prediction is stated by the authors to hold by definition.— “According to the MLE rule, the reliability of the combined multisensory estimate is the sum of each unisensory reliability and is therefore by definition more reliable” | limitations[3] |
| limitation | The review adopts a deliberately broad definition and names what it thereby encompasses.— “This view is implicit to the definition of multisensory or cross-modal integration as any process in which information across sensory modalities is combined to make a perceptual inference (Bizley et al., 2016). This definition encompasses functions which both precede and incorporate cue integration” | limitations[4] |
| limitation | The species scope has an escape clause based on a judgement of standing.— “We will refer to studies with cats and non-human primates if no rodent data are available in the literature, or if results have been seminal to the field” | limitations[5] |
finding · quantity · variable · definition · key-term — grounded in the full text| finding | The review concludes that different multisensory processes reuse the same overlapping anatomical network but are carried out by several distinct micro-circuits running in parallel, each recruited according to stimulus features and behavioral constraints.— “We conclude that distinct multisensory processes share overlapping anatomical substrates, are implemented by diverse neuronal micro-circuitries that operate in parallel, and are flexibly recruited based on factors such as stimulus features and behavioral constraints.” | content_tags.findings[0] |
| finding | Multisensory processing is not one operation but a family of functions that includes segregating modalities as well as integrating them.— “In this review, we expand the framework on the circuit-level mechanisms of cortical multisensory processing by highlighting that multisensory processing is a family of functions – rather than a single operation – which involves not only the integration but also the segregation of modalities.” | content_tags.findings[1] |
| finding | The outcome of multisensory processing is set jointly by physical stimulus properties (bottom-up) and by cognitive resources and situational demands (top-down).— “We pursue our main aim by considering a scheme in which the outcome of multisensory processing is dependent on both physical stimulus properties (low level factors), cognitive resources and situational demands (high level factors)” | content_tags.findings[2] |
| finding | The same anatomical circuit, such as direct V1-A1 connections, can exert different effects depending on the type of stimulus, implementing distinct cross-modal behaviours.— “Thus, the same anatomical circuit in the primary sensory cortex (e.g. direct connections between V1 and A1) can harbor distinct connection patterns to exert different effects based for instance on the type (volume, frequency, temporal pattern, cross-modal synchrony) of sensory stimuli, and thus implement distinct forms of context-dependent… | content_tags.findings[3] |
| finding | The two population-level mechanisms previously proposed (divisive normalization and phase resetting) cannot account for the whole of multisensory processing; a larger set of computational rules is needed.— “However, while some forms of MP (e.g. cue integration) might be fully explained by the proposed two mechanisms, we believe that a complex, diverse phenomenon such as MP can only be explained by considering a larger set of computational rules.” | content_tags.findings[4] |
| finding | The authors argue that different circuits evolved for multi- and unisensory processing, and multisensory integration is not simply extra converging input.— “This indicates that different circuits evolved for multi- versus unisensory processing, and that multisensory integration cannot just be seen as a simple consequence of additional inputs converging onto multisensory neurons.” | content_tags.findings[5] |
| finding | Correlating unisensory responsiveness with multisensory modulation can produce a spurious inverse-effectiveness result because the two variables are not independent and are subject to regression to the mean.— “First, computing the correlation between two variables which are not independent from one another might suffer from a ‘regression to the mean’ effect.” | content_tags.findings[6] |
| finding | The authors prescribe that any observed inverse-effectiveness correlation be tested against a null distribution built by shuffling before it is claimed as real.— “First, to control for the possibility of regression to the mean, the observed correlation between unisensory responsiveness and multisensory response modulation should be tested against a null-distribution which is obtained by a shuffling procedure.” | content_tags.findings[7] |
| finding | Using only simple cross-modal cues biases the field toward primary sensory cortices and hides the contribution of association cortices.— “Thus, using very simple cues in cross-modal paradigms may lead to a biased view on the involved brain areas and neuronal mechanisms, because involvement of the primary sensory cortices may be… | content_tags.findings[8] |
| finding | Experience may switch cross-modal object recognition from a feedforward route (PPC to perirhinal cortex) to a feedback route in which perirhinal multisensory representations affect earlier sensory cortices. | content_tags.findings[9] |
| finding | Conscious experience is proposed to be a higher-order form of multisensory integration that sits above cue integration, object recognition and sensory selection, none of which alone suffices for consciousness. | content_tags.findings[10] |
| finding | Cue-integration models fit the mice's detection behaviour significantly better than a model of independent processing. | content_tags.findings[11] |
| finding | Rodents can integrate stimuli supra-linearly, better than Bayes-optimal. Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.findings[12] |
| quantity | 80 ms — temporal separation beyond which two cross-modal signals are likely to be treated as coming from different sources and segregated Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.quantities[0] |
| quantity | 0–100 ms — the lead time by which a somatosensory stimulus must precede a visual one for layer-4 responses in rat parietal cortex to be enhanced rather than suppressed Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.quantities[1] |
| quantity | 100% — the misleadingly large percentage increase produced by one extra spike when the baseline response is a single spike, used to argue against percentage metrics of multisensory modulation Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.quantities[2] |
| quantity | p < 0.01 — significance of the drop in novel-object exploration ratio after perirhinal-parietal disconnection lesions in cross-modal object recognition Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.quantities[3] |
| variable | stimulus intensity (independent) Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.variables[0] |
| variable | Correct responses (dependent) Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.variables[1] |
| definition | multisensory cue integration Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.definitions[0] |
| definition | sensory selection Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.definitions[1] |
| definition | multisensory Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.definitions[2] |
| definition | principle of inverse effectiveness Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.definitions[3] |
| definition | temporal principle of multisensory integration Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.definitions[4] |
| definition | unisensory integration Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.definitions[5] |
| definition | Maximum-Likelihood Estimate Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.definitions[6] |
| definition | divisive normalization Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.definitions[7] |
| definition | association cortices Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.definitions[8] |
| concept | multisensory processing Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | concepts[0] |
| concept | segregation Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | concepts[1] |
| concept | integration Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | concepts[2] |
| concept | cue integration Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | concepts[3] |
| concept | MLE rule Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | concepts[4] |
| model fitting / model comparison | as written: assessing how well models of cue-combination based on signal detection theory fitted to our behavioral data Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.statistics[0] |
| p-value p | as written: ** p < 0.01 Quotation withheld — this paper is not ours to reproduce, and the page has already quoted as much of it as the rights layer allows. The tag, its field and its confidence stand; the sentence is at the publisher. | content_tags.statistics[1] |
When we have changed our mind. Some entries in the table begin with a word in capitals and the reason for it. EVIDENCE CORRECTED — the finding still stands, but we had quoted the wrong sentence from the paper, and the quotation has been replaced with the right one. RETRACTED — the entry itself was wrong; it stays on the page, marked, with the reason, rather than being deleted. WITHDRAWN AND REPLACED — the entry was wrong, and a corrected version of it appears in the entries below. Each mark carries its own reason, in brackets, because the reasons differ from one entry to the next. An entry we have WITHDRAWN is removed from this page and kept in the record, with its reason. A line reading Audit and a date is a later re-reading of the entry against its whole sentence: DOUBTFUL means it was kept, and the reason says what is loose about it. Quotation marks inside the tagger's sentence. Where a tag is the tagger's own sentence, a phrase in quotation marks inside it is borrowed from the paper — the tagger's wording around it is not. The paper's own sentence is the evidence quote for that row; on this page it is withheld with the article, and the article is at its publisher. Nothing inside quotation marks anywhere on this page is ours: if it is quoted, it is the paper's. Whose words a concept carries. A concept entry records a term the paper itself uses, so the term, the sense given for it, and its evidence are the paper's own words, quoted where the paper fixes them. This is the opposite of a sense entry in the tag table, which is the tagger's sentence about how the paper uses a term. Where a concept entry could not be grounded in the paper's words, the entry says so on its own row.
Tag layer © the Meta-Tagging Project, CC BY 4.0. The paper text is © its authors/publisher
and is linked, not reproduced. Excerpts shown are short quotations for identification and verification.
· back to index
· concepts · authors
· עברית
© 2026 Shir Sivroni · ORCID 0009-0000-2597-2824