Joint Attention Research Articles

Highlighting as a referential and collaborative practice in multiparty video-mediated learning activities

ABSTRACT This paper investigates the accomplishment of highlighting in multiparty video-mediated learning activities. The focus is on moments of screen sharing and the verbal, embodied, and digital resources participants deploy to draw joint attention to a referent on screen. Drawing on screen recorded data (90 h) from an adult education setting (i.e. online crisis management course) and using multimodal conversation analysis (CA), the paper illustrates that highlighting consists of carefully coordinated conduct, including the mobilisation of the cursor and the selection of parts of text in concert with talk and bodily-visual practices. The analysis shows highlighting as a referential and collaborative practice used: 1) to establish and maintain co-participants’ attention to specific topics and items (e.g. words) during multi-unit turns, and 2) to initiate negotiations of written content. The study contributes to a better understanding of screens and digital objects as collaborative and structuring resources and of how they can be meaningfully deployed in task-based online teaching and learning.

Participant attention on the intervention target during repetitive passive movement improved spinal reciprocal inhibition enhancement and joint movement function

This study aimed to evaluate the effects of the participant’s attention target during repetitive passive movement (RPM) intervention on reciprocal inhibition (RI) and joint movement function. Twenty healthy adults participated in two experiments involving four attention conditions [control (forward attention with no RPM), forward attention (during RPM), monitor attention (monitor counting task during RPM), ankle joint attention (ankle movement counting task during RPM)] during 10-min RPM interventions on the ankle joint. Counting tasks were included to ensure the participant’s attention remained on the target during the intervention. In Experiment 1, RI was measured before, immediately after, and 5, 10, 15, 20, and 30 min after the RPM intervention. In Experiment 2, we evaluated ankle joint movement function at the same time points before and after RPM intervention. The maximum ankle dorsiflexion movement (from 30° plantar flexion to 10° dorsiflexion) was measured, reflecting RI. In Experiment 1, the RI function reciprocal Ia inhibition was enhanced for 10 min after RPM under all attention conditions (excluding the control condition. D1 inhibition was enhanced for 20 min after RPM in the forward and monitor attention conditions and 30 min after RPM in the ankle joint attention condition. In Experiment 2, the joint movement function decreased under the forward and monitor attention conditions but improved under the ankle joint attention condition. This study is the first to demonstrate that the participant’s attention target affected the intervention effect of the RI enhancement method, which has implications for improving the intervention effect of rehabilitation.

What We Do and Don’t Know About Joint Attention

Joint attention is an early-emerging and uniquely human capacity that lies at the foundation of many other capacities of humans, such as language and the understanding of other minds. In this article, I summarize what developmentalists and philosophers have come to find out about joint attention, and I end by stating that two problems or questions of joint attention require additional research: 1) the relation between joint attention and the skills for dyadic sharing or affect exchange in young infants, and 2) the evolution of joint attention.

Developmental Play Skills as Outcomes of Early Intervention.

Play is critical in the development of cognitive and language skills in young children with autism; however, few studies have examined the impact of the intervention on the development of play skills. The current study aims to address the change in developmental skills, including play after participation in the JASPER intervention, and to examine the initiation of joint attention as a moderator of the effect of treatment on changes in play skills. The study included 109 preschool-age children who participated in the JASPER intervention and examined changes in developmental skills, including play skills. Children who received JASPER improved significantly in both play diversity (F(1,99) = 4.89, p = 0.029, ES = 0.22) and complexity (F(1,98) = 5.21, p = 0.025; ES = 0.23) compared to children in control conditions. These gains in play skills were associated with concurrent improvements in cognition and communication skills. Additionally, children with more initiations of joint attention skills at entry made greater progress in play diversity (F(1,97) = 15.85, p < 0.001 ES = 0.40) and complexity (p = 0.096). Play and joint attention skills are critical intervention targets and outcomes for children with autism.

Emergent communication through Metropolis-Hastings naming game with deep generative models

Constructive studies on symbol emergence systems seek to investigate computational models that can better explain human language evolution, the creation of symbol systems, and the construction of internal representations. Specifically, emergent communication aims to formulate a computational model that enables agents to build efficient sign systems and internal representations. This study provides a new model for emergent communication, which is based on a probabilistic generative model (PGM) instead of a discriminative model based on deep reinforcement learning. We define the Metropolis-Hastings (MH) naming game by generalizing previously proposed models. It is not a referential game with explicit feedback, as assumed by many emergent communication studies. Instead, it is a game based on joint attention without explicit feedback. Mathematically, the MH naming game is proved to be a type of MH algorithm for an integrative PGM that combines two agents that play the naming game. From this viewpoint, symbol emergence is regarded as decentralized Bayesian inference, and semiotic communication is regarded as inter-personal cross-modal inference. This notion leads to the collective predictive coding hypothesis regarding language evolution and, in general, the emergence of symbols. We also propose the inter-Gaussian mixture model (GMM)+ variational autoencoder (VAE), a deep generative model for emergent communication based on the MH naming game. In this model, two agents create internal representations and categories and share signs (i.e. names of objects) from raw visual images observed from different viewpoints. The model has been validated on MNIST and Fruits 360 datasets. Experimental findings demonstrate that categories are formed from real images observed by agents, and signs are correctly shared across agents by successfully utilizing both of the observations of agents via the MH naming game. Furthermore, scholars verified that visual images were recalled from signs uttered by agents. Notably, emergent communication without supervision and reward feedback improved the performance of the unsupervised representation learning of agents.

A Two-Stage Three-Dimensional Attention Network for Lightweight Image Super-Resolution

In recent years, single image super-resolution (SISR) methods using convolutional neural networks (CNN) have achieved satisfactory performance. Nevertheless, the large model scale and the slow inference speed of these methods greatly limit the application scenarios. In this paper, we propose a two-stage three-dimensional attention network (ATTNet) for lightweight image super-resolution. First, we put forward the spatial feature encoder–decoder (SFE-D) with a spatial attention mechanism. Next, the channel transposed attention module (CTAM) with a channel self-attention mechanism is designed. Both the modules are used for fine feature extraction in the low-resolution stage. Finally, the content-based pixel recombination module (CPRM) is proposed to reconstruct the detailed content with a joint attention mechanism in the high-resolution stage. According to our experimental results, significant performance in terms of the quantitative metrics and the subjective visual quality can be achieved on average compared with the state-of-the-art lightweight SISR algorithms.

Machine learning for distinguishing saudi children with and without autism via eye-tracking data

BackgroundDespite the prevalence of Autism Spectrum Disorder (ASD) globally, there’s a knowledge gap pertaining to autism in Arabic nations. Recognizing the need for validated biomarkers for ASD, our study leverages eye-tracking technology to understand gaze patterns associated with ASD, focusing on joint attention (JA) and atypical gaze patterns during face perception. While previous studies typically evaluate a single eye-tracking metric, our research combines multiple metrics to capture the multidimensional nature of autism, focusing on dwell times on eyes, left facial side, and joint attention.MethodsWe recorded data from 104 participants (41 neurotypical, mean age: 8.21 ± 4.12 years; 63 with ASD, mean age 8 ± 3.89 years). The data collection consisted of a series of visual stimuli of cartoon faces of humans and animals, presented to the participants in a controlled environment. During each stimulus, the eye movements of the participants were recorded and analyzed, extracting metrics such as time to first fixation and dwell time. We then used these data to train a number of machine learning classification algorithms, to determine if these biomarkers can be used to diagnose ASD.ResultsWe found no significant difference in eye-dwell time between autistic and control groups on human or animal eyes. However, autistic individuals focused less on the left side of both human and animal faces, indicating reduced left visual field (LVF) bias. They also showed slower response times and shorter dwell times on congruent objects during joint attention (JA) tasks, indicating diminished reflexive joint attention. No significant difference was found in time spent on incongruent objects during JA tasks. These results suggest potential eye-tracking biomarkers for autism. The best-performing algorithm was the random forest one, which achieved accuracy = 0.76 ± 0.08, precision = 0.78 ± 0.13, recall = 0.84 ± 0.07, and F1 = 0.80 ± 0.09.ConclusionsAlthough the autism group displayed notable differences in reflexive joint attention and left visual field bias, the dwell time on eyes was not significantly different. Nevertheless, the machine algorithm model trained on these data proved effective at diagnosing ASD, showing the potential of these biomarkers. Our study shows promising results and opens up potential for further exploration in this under-researched geographical context.

A linguistic-pragmatic analysis of cat-induced deixis in cat-human interactions

The present paper contributes to the emerging field of embodied interaction. It reports on research into deictic interactions between a human and a non-human, specifically a cat, interlocutor, applying a pragmatic framework developed for human–human communicative interactions. We analyse video recordings of interactions where a cat is a deictic agent pointing the human interlocutor either to the door or the food bowl. We show that these interactions show triadic pointing, hence, focusing joint attention on a common object as a proxy for the event i.e., providing food in the bowl and opening the door. We show that the cat also checks whether the human understands her intentions and that she confirms the human's interpretation. We do not restrict ourselves to vocal communication - as is often done in human language studies -, but we examine how in cat-human communication the cat and human bodies are used to express deixis. Thus, we conceptualize deixis as an embodied interpersonal i.e., interspecies phenomenon. We show that the cat interlocutor uses her body, e.g. eyes/body/tail/ears, as well as her voice, meowing/purring, within this complex deictic interaction.

Addressing challenges in multi-label text classification: A joint attention and shared semantic space approach

The task of multi-label text classification involves assigning a set of related labels to a given document. However, there are three main problems with this task. Firstly, the joint modeling of label-text and label-label relationships is inadequate. Secondly, the semantic mining of the label itself is insufficient. Lastly, the utilization of the internal structure information of the label is ignored. To address these issues, a new multi-label text classification method has been proposed. This method is based on joint attention and shared semantic space. The joint multi-head attention mechanism models the relationship between labels and documents as well as the relationship between labels simultaneously. This helps to avoid error transmission and utilizes the interaction information between them. The decouple shared semantic space embedding method improves the method of using labels semantic information and reduces deviation in the phase of modeling correlation. The hierarchical hinting method based on prior knowledge relies on the prior knowledge in the pre-trained model to exploit the labels hierarchy information. Experimental results have shown that this proposed method is superior to existing multi-label text classification methods in public datasets.

Effects of Shared Attention on joint language production across processing stages

ABSTRACT Shared attention across individuals is a crucial component of joint activities, modulating how we perceive relevant information. In this study, we explored shared attention in language production and memory across separate representation levels. In a shared go/no-go task, pairs of participants responded to objects displayed on a screen: One participant reacted according to the animacy of the object (semantic task), while her partner reacted to the first letter/phoneme (phoneme-monitoring task). Objects could require a response from either one participant, both participants or nobody. Only participants assigned to the phoneme-monitoring task were faster at responding to the joint than to alone trials. However, results from a memory recall test showed that for both partners recall was more accurate for those items to which the partner responded and for jointly responded items. Overall, our findings suggest that partners co-represent each other’s language features even when they do not engage in the same task.

Beyond learning opportunities: focused encounters in a sociocognitive approach to second language acquisition and teaching

Abstract This paper introduces the sociological concept of focused encounters (FEs) as a tool for researching interaction in second language acquisition and teaching. FEs are face-to-face interactions affording joint attention, intersubjectivity, interaffectivity, and interactional alignment, considered key contributors to learning in some SLA/T approaches. First, we introduce the FE concept. Second, we situate FEs within a sociocognitive approach to SLA/T. Third, we apply the FE concept in investigating interaction in an Australian English as a second language classroom. Fourth and finally, we discuss our analysis’s wider implications for learning and teaching: Studying interaction in FEs from a sociocognitive perspective enables investigation not just of language learning opportunities but of moment-by-moment learning processes.

Decreased frontal lobe complexity in left-behind children during joint attention: a fNIRS study with multivariable and multiscale sample entropy analysis.

Human brain development is shaped by experiences, especially during preschool, the critical period for cognitive and socioemotional development. This study employed the functional Near-Infrared Spectroscopy technique to explore the neural differences between left-behind children (LBC) and non-left-behind children (NLBC) on joint attention. Through collecting brain image data of 50 children (26 boys, aged 65.08 ± 6.28months) and conducting multivariable and multiscale sample entropy (MMSE) analysis, the present study found that: (i) LBC showed lower brain complexity than NLBC in right prefrontal cortex; (ii) all participants demonstrated higher brain complexity in responding to joint attention conditions, compared to initiating joint attention ones; (iii) their brain complexity during joint attention was negatively associated with their emotional abilities. The findings advance our understanding of early brain development in LBC by providing evidence for the neural process characteristics of joint attention. Implications for early intervention to promote their brain development are also addressed.

Effect of race on Gaze Cueing in adults with high and low autistic traits

BackgroundObserving the direction of gaze of another person leads to shifting of attention in the same direction (gaze-cueing effect – GCE), a social-cognitive ability known as joint or social attention. Racial attitudes can influence the magnitude of GCE since it has been shown that White people showing a strong race ingroup preference follow the gaze only of White, and not Black, faces. Individuals with high autistic traits have difficulties in social-cognitive abilities that can disrupt the learning of socially shared racial attitudes. Our aim was to investigate in White Italian adults whether individuals with higher autistic traits (measured by the Autism Spectrum Quotient) show reduced implicit racial bias (measured by the Implicit Association Test) and if this bias would lead to differences in the gaze cueing effect (GCE) triggered by gaze direction of faces of different races (measured by the Gaze Cueing Task).MethodsIn an online study, participants (N = 165; 132 females; Mean age = 22.9; SD = 4.76) filled in the Autism Spectrum Quotient (AQ) questionnaire, then performed a Gaze Cueing Task, followed and by an Implicit Association Test.ResultsLinear regression and linear mixed model analyses showed in the IAT task the presence of the same implicit ingroup bias for all participants, which was not predicted by the AQ score, while in the Gaze Cueing Task the GCE differed depending on the AQ score of the participants. Specifically, participants with low-medium, medium, and medium–high autistic traits (AQ = -1SD; AQ = mean; AQ = + 1SD respectively) presented the GCE for both ingroup and outgroup cueing faces, whereas participants with high autistic traits (AQ = + 2SD) only for ingroup faces.ConclusionsIn White Italian adults the presence of an implicit ingroup bias seems to influence the GCE, but it is not always true that the individuals showing an implicit ingroup bias do not orient their attention in the direction of gaze of the outgroup individuals. Instead, the GCE seems to be modulated by the level of autistic traits. That is, individuals with higher autistic traits seem to prioritize joint attention with only their ingroup members.

The limits of thematization

AbstractA fundamental capacity of language is its reflexivity. But not every aspect of language is equally accessible to being reflected upon. Michael Silverstein's 1981 paper, the “Limits of Awareness,” set the terms of this discussion in linguistic anthropology with his study of speakers' “awareness” of pragmatic forms and their corresponding capacity to talk about them. His notion of differential “awareness” of aspects of language has since been foundational to linguistic‐anthropological understandings of language ideologies. Here we consider Silverstein's argument with reference to our research in Laos, exploring the limits of metalinguistic discourse. We argue that the apparent constraints on our capacity to talk about aspects of language do not evidence limits of awareness of elements of language, but rather constraints on our ability to thematize those elements, that is, to bring them into joint attention. The central issue is thematization, and the relation of interest is a relation of joint attention between speakers. Metalanguage is thus constrained not (only) by psychological limits but by the social and semiotic limits on what people can bring into mutual focus within interactions. To present our framing of the issue and show what it helps us see, we distinguish two kinds of thematization and describe their subtypes, affordances, and constraints. We then demonstrate how social conventions—broadly understood—can circumvent these constraints, allowing people to thematize otherwise difficult to thematize forms.

Efficient Net-based Transfer Learning Technique for Facial Autism Detection

Autism Spectrum Disorder is a neurological disorder in which an individual faces life-long effects in communication and interaction with others. Nowadays, the Autism Spectrum disorder ratio is increasing drastically more than ever before. Autism can be identified at all developmental levels as a ”behavioural condition,” and its symptoms often arise between the ages of two and four. The ASD issue starts during puberty and persists through adolescence and adulthood. Children with ASD use both nonverbal and verbal behaviour to communicate, and they struggle with joint attention and social reciprocity. Children with autism are frequently socially isolated as a result of these problems. Through very expensive and time-consuming screening exams, autism spectrum features can be identified. As one of the possible mirrors of the brain, children’s faces can be utilised as a biomarker and as a quick and convenient technique for the early identification of ASD. An effective, genuine, and automatic method of face-based spectrum disorder identification is required. In this study we compare the transfer learning approach used for autism identification with the convolutional neural network (CNN)-based efficient-net strategy to identify autistic children using facial images. We used an open-source Kaggle dataset and evaluated the model performance in terms of accuracy, confusion matrix, precision, recall, and F1 measure. Efficient shows an accuracy of 97% on the benchmark dataset and beats the baseline technique of transfer learning-based approaches. This study can be used to help medical professionals validate their initial screening procedures for finding youngsters with ASD disease.

Is there continuity from implicit recognition of intentional action in infants to explicit mindreading in preschoolers? Systematic review of longitudinal evidence and theoretical implications (¿Hay continuidad entre el reconocimiento implícito de la acción intencional en bebés y la lectura de mente explícita en preescolares? Revisión sistemática de la evidencia longitudinal e implicaciones teóricas)

ABSTRACT The idea that implicit and explicit mindreading form a continuum is controversial. This paper presents a systematic review of longitudinal findings on the development of mindreading in children to compare the main theoretical explanations of this capacity (enrichment and conceptual change theories, submentalizing theory and two-system theory). The longitudinal findings on the predictive relationships between implicit and explicit measures of mindreading demonstrate a continuum between children’s implicit and explicit capacities to understand others’ mental states. Moreover, the findings support the theory of continuous conceptual enrichment and change and provide evidence against submentalist theories. This paper proposes that implicit mindreading is based on the interactive processes of joint attention and shared intentionality that precede explicit mind comprehension.

Autism Spectrum Disorders in Children

Background: Autism Spectrum Disorders (ASDs) describe a category of neurodevelopmental disorders in which individuals experience difficulties of social interaction and age-appropriate play and struggle to establish healthy peer interactions at their stage of development. This study will address the various aspects of Autism in children including clinical symptoms, evaluation, diagnosis and widely used medication or care. Methodology: The search strategy involved utilizing two primary sources for obtaining relevant literature: Google Scholar and PubMed. The inclusion criteria comprised articles published in peer-reviewed journals, written in English, and focusing on ASD in children. Studies with diverse designs, including experimental, observational, and review studies, were considered. Results: While children with autism spectrum disorders share many characteristics with children that have other developmental disorders and can benefit from many of the same educational strategies, they pose specific challenges for families, teachers, and others who collaborate with them. A deficiency in non-verbal and verbal communication requires a great deal of effort and ability, including in teaching basic knowledge. Special challenges in social interaction (e.g.,joint attention) may require more individual instruction than most children to gain and retain their children’s attention. Furthermore, ordinary social interactions between peers do not typically take place without active preparation and ongoing adult structuring in the child’s environment. Conclusion: The lack of traditional friendships and peer relationships influences child motivation structures and the sense of experience. Adequate social experiences can be some of the most challenging and valuable lessons a child with autism spectrum disorders can learn. Keywords: ASD (Autism Spectrum Disorders), QOL (Quality Of Life), neurodevelopmental conditions, PA (Physical Activity), DSM, ICD

The Novel Instance Segmentation Method Based on Multi-Level Features and Joint Attention

The Novel Instance Segmentation Method Based on Multi-Level Features and Joint Attention

Face Super-Resolution via Joint Edge Information and Attention Aggregation Network

Face super-resolution (SR) is a specialized image super-resolution problem that is particularly relevant in various intelligent transportation scenes such as video surveillance and identification systems. Deep learning-steered solutions for face images often leverage facial priors (i.e., face parsing and landmark) to restore intricate facial components and have demonstrated promising performance. However, these methods typically rely on abundant manually labeled data and require lengthy training times. In this work, we introduce a meaningful approach called the Joint Edge Information and Attention Aggregation Network (JEANet) for the face SR problem. JEANet is established on our carefully designed Face Attention Aggregation Module (FAAM), which incorporates parallel connections of both channel-wise and spatial-wise features. Specifically, we integrate an attention fusion strategy into the residual blocks and elaborately employ edge blocks to extract edge information. Additionally, we introduce adaptive shortcuts that interpolate reconstruction parts at multiple scales. With the aid of adaptive shortcuts, JEANet effectively restores detailed images textures by leveraging contextual features optimally. These considerations enable the convolutional operations to flexibly distill information related to primary facial structures and progressively incorporate edge information to assemble local and global meaningful structures. Qualitative and quantitative evaluations demonstrate that our proposed method excels in recovering highly realistic face images compared to several competitive face SR methods.

A novel bearing fault diagnosis method based joint attention adversarial domain adaptation

Recently, many unsupervised domain adaptation methods based on a metric distance or adversarial training do not consider whether the feature representations can be transferred or not. To overcome this challenge, we explore developing a novel approach named joint attention adversarial domain adaptation (JAADA). Specifically, the extracted features are first manually divided into numbers of feature regions. Second, MMD is introduced to mitigate the distribution discrepancy in separated segment features. Furthermore, different weights obtained by the attention mechanism and MMD values have been assigned to different regions. Finally, local and global attention has been fused into one unified adversarial domain adaptation framework. A series of comprehensive experiments on four fault datasets validate that the proposed method has a superior convergence and could boost 1.9%, 3.0%, 2.1%, and 3.5% accuracy than the state-of-the-art methods, respectively.