The nature of magnitude integration: Contextual interference versus active magnitude binding

Abstract

Magnitude dimensions such as duration and numerosity have been shown to systematically interact, biasing each other in a congruent fashion: the more numerous a set of items is, the longer it is perceived to last in time. This integration between dimensions plays an important role in defining how we perceive magnitude. So far, however, the nature of magnitude integration remains unclear. Is magnitude integration a contextual interference, occurring whenever different types of information are concurrently available in the visual field, or does it involve an active “binding” of the different dimensions of the same object? To address these possibilities, we measured the integration bias induced by numerosity on perceived duration, in two cases: with duration and numerosity conveyed by distinct stimuli, or by the same stimulus. We show that a congruent integration effect can be observed only when the two magnitudes belong to the same stimulus. Instead, when the two magnitudes are conveyed by distinct stimuli, we observed an opposite effect. These findings demonstrate for the first time that a congruent integration occurs only between the dimensions of the same stimulus, suggesting the involvement of an active mechanism integrating the different dimensions of the same object in a unified percept.

Near-optimal integration of the magnitude Information of time and numerosity

“…As discussed above, magnitude interaction is qualitatively consistent with the MLE prediction that each piece of information is integrated based on its reliability (14)(15)(16)(17)(18).…”

“…Overall, the results of both the present and previous studies suggest that magnitude interaction is caused by an adaptive mechanism that flexibly changes with the reliability and time course of a stimulus. Furthermore, it may reflect an integration mechanism to obtain a unified representation of the stimulus magnitude (17) rather than an interference effect (16,18). However, a caveat to our study is that we used a different experimental paradigm from the one often used to study magnitude interaction.…”

“…Interestingly, a recent study suggests that magnitude interaction reflects an active "binding" mechanism that integrates different magnitudes of the same stimulus to achieve a unified magnitude representation (17), rather than a contextual interference . CC-BY 4.0 International license perpetuity.…”

“…It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for this this version posted February 3, 2023. ; https://doi.org/10.1101/2023.02.01.526584 doi: bioRxiv preprint mechanism as was conventionally assumed (5,10,18,19). This raises the possibility that magnitude interaction (also referred to as "magnitude integration" (3,17)) is considered one form of "cue integration" (20)(21)(22). Cue integration typically occurs when a single physical quantity is specified using multiple features that provide redundant information.…”

“…According to a tutorial on cue integration research, 10 participants should be sufficient to test the MLE model (22). However, because previous studies on magnitude integration have tested approximately 15 to 20 participants (2,3,17), we recruited 21 participants for each of Experiments 1 and 2.…”