Explore the vast range of titles available.

The brain efficiently processes multisensory information by selectively combining related signals across the continuous stream of multisensory inputs. To do so, it needs to detect correlation, lag and synchrony across the senses; optimally integrate related information; and dynamically adapt to spatiotemporal conflicts across the senses. Here we show that all these aspects of multisensory perception can be jointly explained by postulating an elementary processing unit akin to the Hassenstein–Reichardt detector—a model originally developed for visual motion perception. This unit, termed the multisensory correlation detector (MCD), integrates related multisensory signals through a set of temporal filters followed by linear combination. Our model can tightly replicate human perception as measured in a series of empirical studies, both novel and previously published. MCDs provide a unified general theory of multisensory processing, which simultaneously explains a wide spectrum of phenomena with a simple, yet physiologically plausible model. The human brain integrates inputs across multiple sensory streams into a unified percept. Here Parise and Ernst present a model that assesses the correlation, lag and synchrony of multisensory stimuli, and predicts psychophysical performance on multisensory temporal and spatial tasks.

Human perception, cognition, and action are laced with seemingly arbitrary mappings. In particular, sound has a strong spatial connotation: Sounds are high and low, melodies rise and fall, and pitch systematically biases perceived sound elevation. The origins of such mappings are unknown. Are they the result of physiological constraints, do they reflect natural environmental statistics, or are they truly arbitrary? We recorded natural sounds from the environment, analyzed the elevation-dependent filtering of the outer ear, and measured frequency-dependent biases in human sound localization. We find that auditory scene statistics reveals a clear mapping between frequency and elevation. Perhaps more interestingly, this natural statistical mapping is tightly mirrored in both ear-filtering properties and in perceived sound location. This suggests that both sound localization behavior and ear anatomy are fine-tuned to the statistics of natural auditory scenes, likely providing the basis for the spatial connotation of human hearing.

Neural mechanisms that arbitrate between integrating and segregating multisensory information are essential for complex scene analysis and for the resolution of the multisensory correspondence problem. However, these mechanisms and their dynamics remain largely unknown, partly because classical models of multisensory integration are static. Here, we used the Multisensory Correlation Detector, a model that provides a good explanatory power for human behavior while incorporating dynamic computations. Participants judged whether sequences of auditory and visual signals originated from the same source (causal inference) or whether one modality was leading the other (temporal order), while being recorded with magnetoencephalography. First, we confirm that the Multisensory Correlation Detector explains causal inference and temporal order behavioral judgments well. Second, we found strong fits of brain activity to the two outputs of the Multisensory Correlation Detector in temporo-parietal cortices. Finally, we report an asymmetry in the goodness of the fits, which were more reliable during the causal inference task than during the temporal order judgment task. Overall, our results suggest the existence of multisensory correlation detectors in the human brain, which explain why and how causal inference is strongly driven by the temporal correlation of multisensory signals. Neural mechanisms that arbitrate between integrating and segregating multisensory information are essential for complex scene analysis. Here, the authors show the existence of multisensory correlation detectors in the human brain which explains why and how causal inference is driven by the temporal correlation of multisensory signals.

Audiovisual information reaches the brain via both sustained and transient input channels, representing signals’ intensity over time or changes thereof, respectively. To date, it is unclear to what extent transient and sustained input channels contribute to the combined percept obtained through multisensory integration. Based on the results of two novel psychophysical experiments, here we demonstrate the importance of the transient (instead of the sustained) channel for the integration of audiovisual signals. To account for the present results, we developed a biologically inspired, general-purpose model for multisensory integration, the multisensory correlation detectors, which combines correlated input from unimodal transient channels. Besides accounting for the results of our psychophysical experiments, this model could quantitatively replicate several recent findings in multisensory research, as tested against a large collection of published datasets. In particular, the model could simultaneously account for the perceived timing of audiovisual events, multisensory facilitation in detection tasks, causality judgments, and optimal integration. This study demonstrates that several phenomena in multisensory research that were previously considered unrelated, all stem from the integration of correlated input from unimodal transient channels.

Our body is made of flesh and bones. We know it, and in our daily lives all the senses constantly provide converging information about this simple, factual truth. But is this always the case? Here we report a surprising bodily illusion demonstrating that humans rapidly update their assumptions about the material qualities of their body, based on their recent multisensory perceptual experience. To induce a misperception of the material properties of the hand, we repeatedly gently hit participants' hand with a small hammer, while progressively replacing the natural sound of the hammer against the skin with the sound of a hammer hitting a piece of marble. After five minutes, the hand started feeling stiffer, heavier, harder, less sensitive, unnatural, and showed enhanced Galvanic skin response (GSR) to threatening stimuli. Notably, such a change in skin conductivity positively correlated with changes in perceived hand stiffness. Conversely, when hammer hits and impact sounds were temporally uncorrelated, participants did not spontaneously report any changes in the perceived properties of the hand, nor did they show any modulation in GSR. In two further experiments, we ruled out that mere audio-tactile synchrony is the causal factor triggering the illusion, further demonstrating the key role of material information conveyed by impact sounds in modulating the perceived material properties of the hand. This novel bodily illusion, the 'Marble-Hand Illusion', demonstrates that the perceived material of our body, surely the most stable attribute of our bodily self, can be quickly updated through multisensory integration.

Sensory information about the state of the world is generally ambiguous. Understanding how the nervous system resolves such ambiguities to infer the actual state of the world is a central quest for sensory neuroscience. However, the computational principles of perceptual disambiguation are still poorly understood: What drives perceptual decision-making between multiple equally valid solutions? Here we investigate how humans gather and combine sensory information-within and across modalities-to disambiguate motion perception in an ambiguous audiovisual display, where two moving stimuli could appear as either streaming through, or bouncing off each other. By combining psychophysical classification tasks with reverse correlation analyses, we identified the particular spatiotemporal stimulus patterns that elicit a stream or a bounce percept, respectively. From that, we developed and tested a computational model for uni- and multi-sensory perceptual disambiguation that tightly replicates human performance. Specifically, disambiguation relies on knowledge of prototypical bouncing events that contain characteristic patterns of motion energy in the dynamic visual display. Next, the visual information is linearly integrated with auditory cues and prior knowledge about the history of recent perceptual interpretations. What is more, we demonstrate that perceptual decision-making with ambiguous displays is systematically driven by noise, whose random patterns not only promote alternation, but also provide signal-like information that biases perception in highly predictable fashion.

Animals excel at seamlessly integrating information from different senses, a capability critical for navigating complex environments. Despite recent progress in multisensory research, the absence of stimulus-computable perceptual models fundamentally limits our understanding of how the brain extracts and combines task-relevant cues from the continuous flow of natural multisensory stimuli. Here, we introduce an image- and sound-computable population model for audiovisual perception, based on biologically plausible units that detect spatiotemporal correlations across auditory and visual streams. In a large-scale simulation spanning 69 psychophysical, eye-tracking, and pharmacological experiments, our model replicates human, monkey, and rat behaviour in response to diverse audiovisual stimuli with an average correlation exceeding 0.97. Despite relying on as few as 0–4 free parameters, our model provides an end-to-end account of audiovisual integration in mammals—from individual pixels and audio samples to behavioural responses. Remarkably, the population response to natural audiovisual scenes generates saliency maps that predict spontaneous gaze direction, Bayesian causal inference, and a variety of previously reported multisensory illusions. This study demonstrates that the integration of audiovisual stimuli, regardless of their spatiotemporal complexity, can be accounted for in terms of elementary joint analyses of luminance and sound level. Beyond advancing our understanding of the computational principles underlying multisensory integration in mammals, this model provides a bio-inspired, general-purpose solution for multimodal machine perception.

A growing body of empirical research on the topic of multisensory perception now shows that even non-synaesthetic individuals experience crossmodal correspondences, that is, apparently arbitrary compatibility effects between stimuli in different sensory modalities. In the present study, we replicated a number of classic results from the literature on crossmodal correspondences and highlight the existence of two new crossmodal correspondences using a modified version of the implicit association test (IAT). Given that only a single stimulus was presented on each trial, these results rule out selective attention and multisensory integration as possible mechanisms underlying the reported compatibility effects on speeded performance. The crossmodal correspondences examined in the present study all gave rise to very similar effect sizes, and the compatibility effect had a very rapid onset, thus speaking to the automatic detection of crossmodal correspondences. These results are further discussed in terms of the advantages of the IAT over traditional techniques for assessing the strength and symmetry of various crossmodal correspondences.

Perception can often be described as a statistically optimal inference process whereby noisy and incomplete sensory evidence is combined with prior knowledge about natural scene statistics. Previous evidence has shown that humans tend to underestimate the speed of unreliable moving visual stimuli. This finding has been interpreted in terms of a Bayesian prior favoring low speed, given that in natural visual scenes objects are mostly stationary or slowly-moving. Here we investigated whether an analogous tendency to underestimate speed also occurs in audition: even if the statistics of the visual environment seem to favor low speed, the statistics of the stimuli reaching the individual senses may differ across modalities, hence potentially leading to different priors. Here we observed a systematic bias for underestimating the speed of unreliable moving sounds. This finding suggests the existence of a slow-motion prior in audition, analogous to the one previously found in vision. The nervous system might encode the overall statistics of the world, rather than the specific properties of the signals reaching the individual senses.

Unlike vision, the mechanisms underlying auditory motion perception are poorly understood. Here we describe an auditory motion illusion revealing a novel cue to auditory speed perception: the temporal frequency of amplitude modulation (AM-frequency), typical for rattling sounds. Naturally, corrugated objects sliding across each other generate rattling sounds whose AM-frequency tends to directly correlate with speed. We found that AM-frequency modulates auditory speed perception in a highly systematic fashion: moving sounds with higher AM-frequency are perceived as moving faster than sounds with lower AM-frequency. Even more interestingly, sounds with higher AM-frequency also induce stronger motion aftereffects. This reveals the existence of specialized neural mechanisms for auditory motion perception, which are sensitive to AM-frequency. Thus, in spatial hearing, the brain successfully capitalizes on the AM-frequency of rattling sounds to estimate the speed of moving objects. This tightly parallels previous findings in motion vision, where spatio-temporal frequency of moving displays systematically affects both speed perception and the magnitude of the motion aftereffects. Such an analogy with vision suggests that motion detection may rely on canonical computations, with similar neural mechanisms shared across the different modalities.