Multimodal Events Research Articles

Together in the Mikrokosmos: Exploring Disabled Embodied Experiences in Designed Sociotechnical Worlds

For geographically dispersed communities, including professional, academic, hobbyist, and fandom communities, in-person events and meetups play a crucial role in the function and health of the whole community. Building off prior literature, we take a nuanced look at large scale community events that blend online and offline interactions across digital and physical embodied experiences in order to build understanding of how this knowledge can be leveraged to build more accessible hybrid modalities for community events. Building on this foundation, this paper presents a comprehensive study using ethnographic methods to examine large scale community events, specifically stadium music concerts of varying modalities (e.g., in-person, live-streamed). Using a disability lens, we explore how such community events are experienced and technologically mediated, as well as how the community and individuals design and construct their own sociotechnical worlds across multiple modalities and embodied experiences. By centering the disabled embodied experience, we examine how community members navigate the concert environment, time, and the social experience of concert attendance. Based on our findings, we highlight the nuanced complexities of inclusive community-centered large scale, multi-modal events.

Optimized Ensemble Approach for Multi-model Event Detection in Big data

Event detection acts an important role among modern society and it is a popular computer process that permits to detect the events automatically. Big data is more useful for the event detection due to large size of data. Multimodal event detection is utilized for the detection of events using heterogeneous types of data. This work aims to perform for classification of diverse events using Optimized Ensemble learning approach. The Multi-modal event data including text, image and audio are sent to the user devices from cloud or server where three models are generated for processing audio, text and image. At first, the text, image and audio data is processed separately. The process of creating a text model includes pre-processing using Imputation of missing values and data normalization. Then the textual feature extraction using integrated N-gram approach. The Generation of text model using Convolutional two directional LSTM (2DCon_LSTM). The steps involved in image model generation are pre-processing using Min-Max Gaussian filtering (MMGF). Image feature extraction using VGG-16 network model and generation of image model using Tweaked auto encoder (TAE) model. The steps involved in audio model generation are pre-processing using Discrete wavelet transform (DWT). Then the audio feature extraction using Hilbert Huang transform (HHT) and Generation of audio model using Attention based convolutional capsule network (Attn_CCNet). The features obtained by the generated models of text, image and audio are fused together by feature ensemble approach. From the fused feature vector, the optimal features are trained through improved battle royal optimization (IBRO) algorithm. A deep learning model called Convolutional duo Gated recurrent unit with auto encoder (C-Duo GRU_AE) is used as a classifier. Finally, different types of events are classified where the global model are then sent to the user devices with high security and offers better decision making process. The proposed methodology achieves better performances are Accuracy (99.93%), F1-score (99.91%), precision (99.93%), Recall (99.93%), processing time (17seconds) and training time (0.05seconds). Performance analysis exceeds several comparable methodologies in precision, recall, accuracy, F1 score, training time, and processing time. This designates that the proposed methodology achieves improved performance than the compared schemes. In addition, the proposed scheme detects the multi-modal events accurately.

Multimodal Event Processing: A Neural-Symbolic Paradigm for the Internet of Multimedia Things

With the Internet of Multimedia Things (IoMT) becoming a reality, new approaches are needed to process real-time multimodal event streams. Existing approaches to event processing have limited consideration for the challenges of multimodal events, including the need for complex content extraction, and increased computational and memory costs. This article explores event processing as a basis for processing real-time IoMT data. This article introduces the multimodal event processing (MEP) paradigm, which provides a formal basis for native approaches to neural multimodal content analysis (i.e., computer vision, linguistics, and audio) with symbolic event processing rules to support real-time queries over multimodal data streams using the multimodal event processing language to express single, primitive multimodal, and complex multimodal event patterns. The content of multimodal streams is represented using multimodal event knowledge graphs to capture the semantic, spatial, and temporal content of the multimodal streams. The approach is implemented and evaluated within a MEP engine using single and multimodal queries achieving near real-time performance with a throughput of ~30 frames processed per second (fps) and subsecond latency of 0.075–0.30 s for video streams of 30 fps input rate. Support for high input stream rates (45 fps) is achieved through content-aware load-shedding techniques with a ~127X latency improvement resulting in only a minor decrease in accuracy.

College English Multimodal Teaching Based on Digital Information Technology.

In the era of in-depth application of knowledge technology in the field of education, the teaching mode combining school English teaching and network data technology is widely used in school English education. Teaching resources have become electronic resources under the influence of digital information technology. However, English teaching often ignores all aspects of English ability, so there are still shortcomings in English teaching. The role of information technology makes a single English teaching text model unable to meet the needs of scholars. In recent years, multimodal events have enriched the expression of language input and output content, which is in line with the learning needs of current students. Therefore, under the premise of digital information technology, this paper constructs a multimodal English teaching model from the aspects of multimodal English teaching resources, multimodal English teaching activities, and multimodal English teaching analysis. It aims to mobilize the enthusiasm of scholars to learn English from various aspects and build a multimodal teaching mode of English for teachers supported by digital data technology. Therefore, the experiment shows that the multimodal teaching mode of college English based on digital information technology proposed in this paper will mobilize the enthusiasm of students to learn in a comprehensive way. It greatly improves students' English performance, promotes the development of students' autonomous learning ability and comprehensive language ability, and improves the quality of English teaching by 20%.

Multi-Modal Clustering Events Observed by Horizon-10T and Axion Quark Nuggets

The Horizon-10T collaboration have reported observation of Multi-Modal Events (MME) containing multiple peaks suggesting their clustering origin. These events are proven to be hard to explain in terms of conventional cosmic rays (CR). We propose that these MMEs might be result of the dark matter annihilation events within the so-called axion quark nugget (AQN) dark matter model, which was originally invented for completely different purpose to explain the observed similarity between the dark and the visible components in the Universe, i.e., ΩDM∼Ωvisible without any fitting parameters. We support this proposal by demonstrating that the observations, including the frequency of appearance, intensity, the spatial distribution, the time duration, the clustering features, and many other properties nicely match the emission characteristics of the AQN annihilation events in atmosphere. We list a number of features of the AQN events which are very distinct from conventional CR air showers. The observation (non-observation) of these features may substantiate (refute) our proposal.

Facing #emergencies. The linguistic role of keywords, hashtags and retweets in communicating critical events

In recent years, social media platforms have had a tremendous impact on the online world due to their effectiveness in multimodal communication events (Herring 2001). Social media users benefit from the digital nature of such interactions to gather data for different purposes, including discourse-related ones (Zappavigna 2012). Hashtags, in this sense, have proved to be an effective tool that is used to broaden communication but also to prompt real-life actions, especially to face and manage critical or emergency situations (Olteanu et al . 2015). This specific device is more likely to be used effectively on Twitter, a popular micro-blog used as an information aggregator and catalyst for action (Zappavigna 2015). Following previous studies focusing on the same topic (Burnap et al. 2014; Hughes, Palen 2009), the paper examines examples of context-based words and/or purpose-specific hashtags to explore their use by different sets of users with diverse intentions and aims. Twitter data are retrieved by means of real-time data mining tools (Brooker et al. 2016) to create relevant keyword- or hashtag-based sample corpora dealing with emergency situations (Aug – Sep 2017: two terror attacks and a natural disaster) which have caused remarkable media exposure. Data from such corpora identify some relevant words used in such situations, grouped according to several variables such as event-related, channel-dependent or sentiment-based criteria. Furthermore, an aggregated analysis is carried out in order to retrieve the most common patterns used to highlight performativity, thus emphasising the role of purpose-specific communication. Finally, a comparison of aggregated corpora with different tool-specific features ( retweets ) highlights the importance of such tool-specific devices in magnifying the range of communication effectiveness occurring in a proper ‘online discourse community’ (Herring 2008).

20: Interdisciplinary, Synergistic, Multimodal Interventions Decrease Ventilator-Associated Events

20: Interdisciplinary, Synergistic, Multimodal Interventions Decrease Ventilator-Associated Events

Gestures, grunts, and words: Development in a dynamic system

Pointing has long been considered influential in language acquisition. Certain pre-linguistic vocal expressions may hold even greater value in addressing the transition to language. The goal of the present study is longitudinal evaluation of early communicative development, addressing the influence of pre-linguistic gestures and vocal expressions. This multiple case study report analyzes longitudinal development in five children from 9 to 16 months of age, a critical language transition period. We include gestures of pointing and extending the hand, with interactive as well as request functions. Gestures, communicative grunts, words, and multimodal events combining gesture with vocal accompaniment comprise the data. Results demonstrate group trends and stark individual differences in children’s use of vocal and gestural modalities, and the influence of grunt communication onset on overall communicative frequency in single and combined communicative events. We imbed this analysis within the broader context of mutually interacting variables in a dynamic system. These results argue for greater attention to vocalization as well as gesture in monitoring children’s approach to language development. Based on the role of communicative grunts demonstrated here, this variable should be further studied in both typical and language-delayed children.

Progression from Feature-Specific Brain Activity to Hippocampal Binding during Episodic Encoding.

The hallmark of episodic memory is recollecting multiple perceptual details tied to a specific spatial-temporal context. To remember an event, it is therefore necessary to integrate such details into a coherent representation during initial encoding. Here we tested how the brain encodes and binds multiple, distinct kinds of features in parallel, and how this process evolves over time during the event itself. We analyzed data from 27 human subjects (16 females, 11 males) who learned a series of objects uniquely associated with a color, a panoramic scene location, and an emotional sound while fMRI data were collected. By modeling how brain activity relates to memory for upcoming or just-viewed information, we were able to test how the neural signatures of individual features as well as the integrated event changed over the course of encoding. We observed a striking dissociation between early and late encoding processes: left inferior frontal and visuo-perceptual signals at the onset of an event tracked the amount of detail subsequently recalled and were dissociable based on distinct remembered features. In contrast, memory-related brain activity shifted to the left hippocampus toward the end of an event, which was particularly sensitive to binding item color and sound associations with spatial information. These results provide evidence of early, simultaneous feature-specific neural responses during episodic encoding that predict later remembering and suggest that the hippocampus integrates these features into a coherent experience at an event transition.SIGNIFICANCE STATEMENT Understanding and remembering complex experiences are crucial for many socio-cognitive abilities, including being able to navigate our environment, predict the future, and share experiences with others. Probing the neural mechanisms by which features become bound into meaningful episodes is a vital part of understanding how we view and reconstruct the rich detail of our environment. By testing memory for multimodal events, our findings show a functional dissociation between early encoding processes that engage lateral frontal and sensory regions to successfully encode event features, and later encoding processes that recruit hippocampus to bind these features together. These results highlight the importance of considering the temporal dynamics of encoding processes supporting multimodal event representations.

Towards a Flexible Deep Learning Method for Automatic Detection of Clinically Relevant Multi-Modal Events in the Polysomnogram.

Much attention has been given to automatic sleep staging algorithms in past years, but the detection of discrete events in sleep studies is also crucial for precise characterization of sleep patterns and possible diagnosis of sleep disorders. We propose here a deep learning model for automatic detection and annotation of arousals and leg movements. Both of these are commonly seen during normal sleep, while an excessive amount of either is linked to disrupted sleep patterns, excessive daytime sleepiness impacting quality of life, and various sleep disorders. Our model was trained on 1,485 subjects and tested on 1,000 separate recordings of sleep. We tested two different experimental setups and found optimal arousal detection was attained by including a recurrent neural network module in our default model with a dynamic default event window (F1 = 0.75), while optimal leg movement detection was attained using a static event window (F1 = 0.65). Our work show promise while still allowing for improvements. Specifically, future research will explore the proposed model as a general-purpose sleep analysis model.

HT-KZ detector system: R&D considerations and prototype performance

Following the experience with the Horizon-10T detector systems, a completely new detector of Extensive Air Showers (EAS) has been designed to be built at Nazarbayev University, Astana, Kazakhstan. The motivation of building HorizonT-Kazakhstan is to study in detail the phenomenon of the multi-modal EAS events that were initially explored by Horizon-8T and 10T detector systems. This detector will have a ns-level resolution of charged particle arrival times and pulse shape resolution. Each detection point is designed to have a total scintillation area approximately equal to 7.7 m2 and consists of three 1.6×1.6 m detectors located in a triangular arrangement for the local determination of the EAS disk arrival direction, and to have ns precision in timing between points. Local arrival direction is important following evidence of large angle differences in the multimodal events seen previously.

Aging Impairs Temporal Sensitivity, but not Perceptual Synchrony, Across Modalities.

Encoding the temporal properties of external signals that comprise multimodal events is a major factor guiding everyday experience. However, during the natural aging process, impairments to sensory processing can profoundly affect multimodal temporal perception. Various mechanisms can contribute to temporal perception, and thus it is imperative to understand how each can be affected by age. In the current study, using three different temporal order judgement tasks (unisensory, multisensory, and sensorimotor), we investigated the effects of age on two separate temporal processes: synchronization and integration of multiple signals. These two processes rely on different aspects of temporal information, either the temporal alignment of processed signals or the integration/segregation of signals arising from different modalities, respectively. Results showed that the ability to integrate/segregate multiple signals decreased with age regardless of the task, and that the magnitude of such impairment correlated across tasks, suggesting a widespread mechanism affected by age. In contrast, perceptual synchrony remained stable with age, revealing a distinct intact mechanism. Overall, results from this study suggest that aging has differential effects on temporal processing, and general impairments with aging may impact global temporal sensitivity while context-dependent processes remain unaffected.

Language Development From an Ecological Perspective: Ecologically Valid Ways to Abstract Symbols

ABSTRACTIn the embodied, situated, enacted and distributed approaches to cognition, the coordinative role of language comes to the fore. Language, with its symbolic properties, arises from a multimodal stream of interactive events and gradually gains power to constrain them in a functional and adaptive way. In this article, we attempt to integrate three approaches to information in cognitive systems to provide a theoretical background to the process of development of language as such a coordinator. Ecological psychology provides an explanation for how any behaviors or events become informative through the process of “tuning” to affordances that control individual and collective behavior. The dynamical approach helps to operationalize this control as a functional reduction of degrees of freedom of individual and collective systems. Cognitive semiotics provides a typology of constraints showing their interrelations: it proposes conditions under which informational controls that function as indices and icons may become symbolic, providing a qualitatively different form of constraint, which can be partially ungrounded from the ongoing stream of multimodal events. The article illustrates the proposed processes with examples from actual parent-infant interaction and points to ways of verifying them in a more quantitative way.

Seeing and Being Seen: The Multimodality of Museum Spectatorship

This article argues that museum visiting and the act of ‘spectatorship’, both of which are often assumed to be ocularcentric, are multimodal events. Anchored in Goffman’s dramaturgy and frame analysis theory, as well as Kress’s multimodal and social semiotic theory of representation and communication, this article presents an apposite interpretative and methodological framework to account for what has not been widely addressed by museum studies; that is, the multimodality of the museum experience. By drawing upon audio-visual excerpts of museum encounters, this analysis brings to the fore the embodied visiting and viewing practices of visitors in museum galleries. Specifically, this article highlights the range of modes of communication and representation, beyond gazing and looking, which are employed, negotiated and regulated within the social context of the visit. The article suggests that visitors’ experiences are embodied and performative interactions with the exhibits and other visitors.Key words: embodiment, multimodality, museums, social interaction, visitors

Deep Multimodal Fusion: A Hybrid Approach

We propose a novel hybrid model that exploits the strength of discriminative classifiers along with the representation power of generative models. Our focus is on detecting multimodal events in time varying sequences as well as generating missing data in any of the modalities. Discriminative classifiers have been shown to achieve higher performances than the corresponding generative likelihood-based classifiers. On the other hand, generative models learn a rich informative space which allows for data generation and joint feature representation that discriminative models lack. We propose a new model that jointly optimizes the representation space using a hybrid energy function. We employ a Restricted Boltzmann Machines (RBMs) based model to learn a shared representation across multiple modalities with time varying data. The Conditional RBMs (CRBMs) is an extension of the RBM model that takes into account short term temporal phenomena. The hybrid model involves augmenting CRBMs with a discriminative component for classification. For these purposes we propose a novel Multimodal Discriminative CRBMs (MMDCRBMs) model. First, we train the MMDCRBMs model using labeled data by training each modality, followed by training a fusion layer. Second, we exploit the generative capability of MMDCRBMs to activate the trained model so as to generate the lower-level data corresponding to the specific label that closely matches the actual input data. We evaluate our approach on ChaLearn dataset, audio-mocap, as well as the Tower Game dataset, mocap-mocap as well as three multimodal toy datasets. We report classification accuracy, generation accuracy, and localization accuracy and demonstrate its superiority compared to the state-of-the-art methods.

Multimodal infant-directed communication: how caregivers combine tactile and linguistic cues.

Both touch and speech independently have been shown to play an important role in infant development. However, little is known about how they may be combined in the input to the child. We examined the use of touch and speech together by having mothers read their 5-month-olds books about body parts and animals. Results suggest that speech+touch multimodal events are characterized by more exaggerated touch and speech cues. Further, our results suggest that maternal touches are aligned with speech and that mothers tend to touch their infants in locations that are congruent with names of body parts. Thus, our results suggest that tactile cues could potentially aid both infant word segmentation and word learning.

Poorer resolution for audiotactile than for audiovisual synchrony detection in cluttered displays.

The brain integrates signals from multiple modalities to provide a reliable estimate of environmental events. A temporally cluttered environment presents a challenge for sensory integration because of the risk of misbinding, yet it also provides scope for cross-modal binding to greatly enhance performance by highlighting multimodal events. We present a tactile search task in which fingertips received pulsed vibrations and participants identified which finger was stimulated in synchrony with an auditory signal. Results showed that performance for identifying the target finger was impaired when other fingers were stimulated, even though all fingers were stimulated sequentially. When the number of fingers vibrated was fixed, we found that both spatial and temporal factors constrained performance, because events occurring close to the target vibration in either space or time reduced accuracy. When tactile search was compared with visual search, we found overall performance was lower in touch than in vision, although the cost of reducing temporal separation between stimuli or increasing the presentation rate was similar for both target modalities. Audiotactile performance benefitted from increasing spatial separation between target and distractors, with a particularly strong benefit for locating the target on a different hand to the distractors, whereas the spatial manipulations did not affect audiovisual performance. The similar trends in performance for temporal manipulations across vision and touch suggest a common supramodal binding mechanism that, when combining audition and touch, is limited by the poor resolution of the underlying unisensory representation of touch in cluttered settings. (PsycINFO Database Record

Research on Analysis of Sports Video Multi-Pattern Fusion

In order to effectively integrate multimodal information and multilayer constraints, we present a unified probabilistic framework for sports video analysis. Based the framework, three instances of statistical models are constructed and compared. Experimental results indicate our method with multimodal fusion processes semantic events in sports video more effectively. With the method based on statistics, we'll discuss further sport video content analysis method fusing multimode information. Semantic events in sport videos are in essence multimodal. In television relay, multimode information is integrative used to present video contents like subtitles, narrator's voices, on-site sounds, camera movements, scenarios and images etc. It is incomplete to analyze only one mode. For more effective analysis of events, it's required to study the analytical method which fuses multiple patterns. On the other hand, semantic events in those videos are not isolated. There's some logical or consequential relationship among them. In previous paper, we discussed event detection and recognition with the use of the contextual relationship based on dynamic Bayes network. Now on that basis, we'll explore how to fuse multimode information, which is a key issue we're facing here (1-2). In recent years, the fusion of multimode information has become a hot topic in the field of sport video analysis (3-5). Firstly we introduce the related work. Most of the multimode fusion analysis methods mentioned in previous literatures considered the detection of an isolated event. Unlike them, we propose to detect many events and analyze comprehensively the association among them. Multi-level analysis methods based on statistics are built on the probabilistic graphical model, such as Hidden Markov Model (HMM), dynamic Bayes network (DBN) and their variants. By combining visual graphical model representation and effective reasoning and learning methods, such solutions made fairly good effects. Xie (6-7) et al. applied hierarchical HMM for unsupervised clustering to discover layered structure of video contents. Differently, we introduced the method based on dynamic Bayes network model to do the same work. Through learning of training samples, we fulfilled the detection and recognition of wonderful events in the football match (8). Based on DBN, we developed a common sport video analysis framework. Compared with previous work mentioned above, our approach can not only integrate multimode information for event detection and also deal with hierarchical relationship among events. Although the two functions were stated in previous papers, they as a whole were not paid

Effect of deceptively aggressive bike pacing on sprint-distance triathlon performance

Introduction : Whilst aggressive bike pacing has been found to impair subsequent run performance during triathlon, this strategy appears to enhance total cycle-run time over the sprint-distance format (Suriano & Bishop, 2010). As such, cycling at the highest sustainable intensity (i.e. isolated TT power output) is suggested as the best strategy to optimise overall performance time in the event. However, it is unclear how an athlete’s expectations, beliefs and perceptions influence the effectiveness of such pacing during multi-modal events, such as triathlon. Taylor & Smith (2013) have suggested that practically meaningful changes in triathlon running may result from deceptive pace manipulation, equivalent to the smallest worthwhile change in performance. With this in mind, deceptively aggressive bike pacing may enable triathletes to maximise their sustainable intensity in this discipline, without the impairments in run performance typically associated with this strategy. However, this is yet to be confirmed by experimental evidence. This study therefore examined the effects of deceptively aggressive bike pacing on sprint- distance triathlon performance. Methods : Having been misinformed that they would be taking part in a reliability and validity study, ten non-elite, competitive male triathletes (mean ± SD: age 36.8 ± 8.9 yrs, mass 76.3 ± 7.2 kg and V, I‡ O 2max 54.3 ± 5.7 ml · kg - 1 · min -1 ) completed eight testing sessions (Figure 1). The first four trials served as familiarisation, after which all athletes completed an isolated 500 kJ (~20 km) cycling time-trial (TT) to establish their maximum sustainable intensity over this distance. Each athlete then completed three separate simulated sprint-distance triathlons (0.75 km swim, 500 kJ bike, 5 km run), the first of which established ‘baseline’ (i.e. personal best) simulated triathlon performance (BL). During the remaining two triathlon trials athletes maintained a cycling power output that was 5% greater than that achieved during BL, before completing the run in as short a time as possible (against an on-screen avatar of BL run performance). However, participants were correctly informed of this aggressive cycling strategy before and during only one of the two trials (Hon). Prior to the alternate trial (Dec), participants were misinformed that their mean cycling power output would equal that of BL, with on-screen feedback manipulated so as to display power output 5% below its true value. Results : Mean performance times for TT and simulated triathlon trials are detailed in Table 1. Although these values suggest a trend for faster run performance during Dec, compared to both BL and Hon, this was only statistically significant versus BL ( p < .05). Similarly, whilst there was a non-significant trend for overall triathlon time to be shorter during Dec than Hon (by ~17 sec), the only statistically significant differences were between each of these trials and BL, which was between 2-3% slower overall than both Dec and Hon. Magnitude-based inferences (Table 2) suggest cycling time and power output were almost certainly better during TT, Dec and Hon, in comparison to BL. Whilst Dec and Hon cycling performances were probably worse compared to that of TT, there were almost certainly no differences between the Dec and Hon cycling. Interestingly, whilst any difference appeared of trivial practical importance, it was more likely that Hon running was meaningfully slower, than faster, compared to BL (i.e. 28:57:15% likelihood of Hon being slower, no different, or faster than BL). On the other hand, Dec running was probably faster than both BL and Hon. In terms of overall triathlon performance, there was almost certainly no difference between Dec and Hon, although both were almost certainly faster compared to BL. Conclusion : An athlete’s expectations, beliefs and perceptions regarding aggressive cycle pacing appear to play an important role in optimising subsequent run and overall performance during sprint-distance triathlon. Future studies should further examine how pacing beliefs and expectations influence performance across multi-modal (i.e. triathlon) or multi-stage (i.e. Tour de France) sporting competitions. In particular, more research is needed so as to better understand the importance of multiple perceptual responses (e.g. RPE and affect) to pacing decisions during successive modes of self-paced exercise.

Intrasensory Redundancy Facilitates Infant Detection of Tempo: Extending Predictions of the Intersensory Redundancy Hypothesis.

Research has demonstrated that intersensory redundancy (stimulation synchronized across multiple senses) is highly salient and facilitates processing of amodal properties in multimodal events, bootstrapping early perceptual development. The present study is the first to extend this central principle of the intersensory redundancy hypothesis (IRH) to certain types of intrasensory redundancy (stimulation synchronized within a single sense). Infants were habituated to videos of a toy hammer tapping silently (unimodal control), depicting intersensory redundancy (synchronized with a soundtrack) or intrasensory redundancy (synchronized with another visual event; light flashing or bat tapping). In Experiment 1, 2-month-olds showed both intersensory and intrasensory facilitation (with respect to the unimodal control) for detecting a change in tempo. However, intrasensory facilitation was found when the hammer was synchronized with the light flashing (different motion) but not with the bat tapping (same motion). Experiment 2 tested 3-month-olds using a somewhat easier tempo contrast. Results supported a similarity hypothesis: intrasensory redundancy between two dissimilar events was more effective than that between two similar events for promoting processing of amodal properties. These findings extend the IRH and indicate that in addition to intersensory redundancy, intrasensory redundancy between two synchronized dissimilar visual events is also effective in promoting perceptual processing of amodal event properties.