Joint attention, the synchronized focus on entities or events between social partners, begins to unfold around the age of 3 months and matures gradually until approximately 18 months in typical development. This phenomenon, observed across various species, is deeply ingrained in both ontogeny and phylogeny. However, it may not be universally evident, particularly among children with autism. Diminished joint attention in infancy serves as an early indicator of autism spectrum disorder (ASD), distinguishing individuals with ASD from their typically developing counterparts and those with other disabilities.
A fundamental skill, joint attention plays a pivotal role in information acquisition, influencing subsequent development across diverse domains for both typically developing individuals and those with ASD. The spectrum of joint attention differences among individuals on the autism spectrum predicts adaptive skills, symptoms, social functioning, linguistic proficiency, and cognitive development.
Despite its significance, limited research has explored joint attention and developmental outcomes at multiple time points, necessitating a deeper understanding of the reciprocal influences between joint attention and social-cognitive development.
Examining longitudinal associations between early joint attention and subsequent development aligns with the social-cognitive theory of joint attention. This theory suggests that the development of joint attention, evolving from simpler social behaviors like face-to-face engagement, signifies an emerging comprehension of othersas intentional agents. This, in turn, scaffolds subsequent symbolic development. The parallel and distributed processing (PDP) model of joint attention, on the other hand, says that it happens when you can better combine information about yourself, other people, and how they relate to an outside object (triadic relations). The distinction lies in the relative importance attributed to understanding another person's mind versus the practice of representing triadic relationships.
Joint attention encompasses two distinct types of behavior that evolve independently and may rely on different neural substrates: response to joint attention (RJA) involving gaze or point following and initiation of joint attention (IJA) involving directing others' attention with eye movements or gestures.
IJA further divides into gestural indication and gaze alternation. Theorists speculate that IJA may pose a more central challenge than RJA in autism, yet recent research suggests both RJA and IJA could be fundamental difficulties in individuals with autism.
Research procedures for all research included in this review have received approval from the appropriate institutional ethics committees, and all studies have been carried out in compliance with the 1964 Declaration of Helsinki. The people who took part in the research that were cited all provided their informed permission before taking part.
RJA might depend on posterior and temporal attentional networks, which develop earlier than the anterior networks typically associated with IJA. While RJA could signify a relatively involuntary response to social cues, IJA may represent a more voluntary sharing of experiences. Notably, RJA has been linked to intention understanding in 3-year-olds with ASD. Interestingly, gaze alternation has been linked to the speed of attention disengagement, suggesting the involvement of non-social attention in IJA in ASD.
Among typically developing infants, RJA and IJA display distinct developmental trajectories. Despite continuity within each domain, associations between IJA and RJA are rarely observed. Assessing RJA is simpler than assessing IJA since it involves presenting stimuli rather than observing child-initiated behaviors. As mentioned earlier, autism may involve a more fundamental disturbance in IJA than RJA. Evidence supporting this idea includes the notion that only IJA may be uniquely human, RJA may be atypical only early in development while IJA remains atypical throughout the lifespan in ASD, IJA may be more closely linked to reward processing than RJA, and IJA may be a better predictor of social outcomes than RJA. However, it's important to note that IJA is not exclusive to humans. As we will explore further, RJA may be impaired across the lifespan in autism, there is limited evidence of stronger connections between IJA and reward processing in ASD, and RJA is also associated with social outcomes.
Joint attention is commonly evaluated in a laboratory setting using structured observational measures designed for children up to approximately 30 months of developmental level. Ideally, assessments of joint attention should take place in diverse settings. In one widely used assessment, the Early Social Communication Scales (ESCS), IJA is categorized into mutually exclusive groups: gestural indication or gaze alternation. RJA in the ESCS is differentiated between proximal (to a book) and distal (to posters), with limited research attention given to RJA proximal. Notably, the ESCS does not vary the subtlety of cues for RJA distal, as children receive overlapping verbal, postural, and gestural cues.
Aligning with the observation that children with ASD may struggle with sparser cues, some measures incorporate varying combinations of cues. For instance, the Autism Diagnostic Observation Schedule-2 provides opportunities for both RJA and IJA, with RJA opportunities presented within a hierarchy of increasing prompts. Parent-checklists or a combination of parent-report and direct observation are also employed in joint attention assessments.
Recent efforts by researchers have led to the development of age-appropriate measures for joint attention in older individuals. Hobson and colleagues utilized shared looks as a measure of IJA in older children and adolescents with ASD, while a measure of RJA suitable for children and adolescents aged 7 to 17 employs six naturalistic prompts.
Analyses of retrospective videos have indicated that diminished dyadic behaviors in the first year of lifeprecede a decline in joint attention in the second year in individuals with autism. This suggests that reduced attention to people during infancy may serve as a precursor to diminished joint attention. However, there are instances of reduced Initiating Joint Attention (IJA) between 6 and 12 months of age in ASD, even when no atypicalities in dyadic interactions are observed. Interestingly, live interaction studies in the first year of life have not reported dyadic atypicalities in prospective investigations of infant siblings of children with autism. Six-month-old infants with ASD exhibited slightly more attention to a face compared to typically developing infants, followed by reduced Responding to Joint Attention (RJA) and gestural indication (but not gaze alternation) at 12 months.
Reduced RJA at 14, 15, and 24 months predicted ASD, and infants diagnosed with ASD at 14 months exhibited reduced RJA, gaze alternation, and gestures. Infants later diagnosed with ASD differed only in gaze alternation at 14 months. Gaze alternation, in particular, emerged as a potent predictor of ASD, outperforming RJA and gestural indication.
While there's no documented predictive link between dyadic interactions and later joint attention, concurrent associations have been observed. Three-to-four-year-olds with autism showed less orientation to both social and non-social stimuli, with decreased concurrent IJA and RJA.
Reduced social orienting was concurrently associated with RJA and IJA in children with ASD. These findings suggest that dyadic atypicalities may contribute to reduced joint attention. However, predictive relationships between social orienting and joint attention haven't been established, and joint attention impairments can occur independently of dyadic impairments.
Verbal and cognitive skills also play a role in joint attention, particularly RJA. Children with ASD may not exhibit reduced RJA if they have a non-verbal mental age above 19 months, a verbal mental age above 47 months, or a non-verbal IQ in the normal range. Some researchers propose bidirectional relations between joint attention and language in ASD, highlighting the absence of RJA deficits in children with higher verbal skills.
Despite the developmental period during which RJA deficits may be apparent in ASD, high-functioning individuals on the spectrum still face challenges initiating joint attention. The Autism Diagnostic Observation Schedule includes both RJA and IJA in its diagnostic criteria for the first module, while the second module (for children with better verbal skills) focuses solely on IJA.
Changes in joint attention across development challenge assertions that only IJA remains a difficulty across the lifespan in autism. This may be due to the fact that most joint attention assessments were designed for toddlers. Age-appropriate tests show that older kids and teens with ASD have problems with both RJA and IJA. This suggests that these problems stay with them throughout their lives when tested with developmentally appropriate tests.
Joint attention not only shows concurrent relationships with other developmental domains but also holds predictive value for future development. Between 12 and 18 months, both Responding to Joint Attention (RJA) and Initiating Joint Attention (IJA) were linked to social responsiveness at 3 years for children with ASD. Gaze alternation at 2 years was associated with social symptoms at 4 years among children with ASD. While researchers in social cognition anticipate associations between joint attention and subsequent theory of mind in autism, prospective relations between the two in ASD have yet to be firmly established.
In early childhood, IJA was found to be connected to later peer engagement, and longitudinal links between RJA in early childhood and cognitive development have been identified in ASD. Early childhood RJA was associated with adult social functioning, social symptoms, and non-verbal communication. The links between RJA and adult outcomes were mostly caused by links between RJA and changes in language and cognitive skills. This supports the idea that joint attention is a key skill that helps people learn other skills.
Joint attention appears to play a crucial role in acquiring social cues relevant for mapping words to objects. Prospective associations between both RJA and IJA and structural language skills are frequently observed among children with ASD. RJA at 14 months was linked to language and cognitive skills between 30 and 36 months, while gaze alternation at 20 months was associated with vocabulary at 42 months among children with ASD. IJA and RJA at 4 years were associated with language skills a year later, and RJA (but not IJA) at around 4 years was linked to expressive language about 8 years later. Predictive connections between gestural indication and syntactic development have also been noted among school-age children with ASD. Future research should consider assessing joint attention and a range of outcomes at multiple developmental points to explore potential bidirectional associations. Given the strong links between joint attention and language, it is crucial to control for linguistic ability when examining the relationship between joint attention and other developmental domains.
It is plausible that Responding to Joint Attention (RJA) and Initiating Joint Attention (IJA) involve distinct yet overlapping mechanisms. Various potential mechanisms underpinning atypical joint attention in autism have been proposed, including atypical reflexive gaze following (with some conflicting evidence), impaired integration of joint attention and affect, reduced recognition of the referential significance of gaze, diminished social motivation, and awareness of the reward value of social interaction, as well as atypicalities in non-social attention. Recent evidence suggests that impairments in reflexive gaze following may contribute to RJA deficits in ASD, while understanding referential intent is more likely linked to developmental levels.
Contrary to consistent findings, there is mixed evidence regarding whether decreased social motivation in infancy leads to joint attention impairments. Although dyadic orienting has been concurrently associated with both RJA and IJA, joint attention deficits often do not precede dyadic difficulties. It is suggested that dyadic difficulties might be linked to joint attention impairments, rather than the other way around, as social behaviors typically become more object-centered around 1 year of age, potentially reducing opportunities for social learning.
Aligned with the Parallel Distributed Processing (PDP) model, non-social attention may play a role in joint attention in ASD. Impairments are observed in both social and non-social orienting in ASD, and a non-social measure of attention, disengagement latency, has been linked to IJA. In typically developing infants, visual attention is related to subsequent IQ and language skills. To learn more about how non-social attention and joint attention change over time and how they affect later development, researchers should look at non-social aspects of attention and joint attention at different stages of development along with social cognition tests.
Investigations into reward sensitivity and joint attention from infancy to early childhood can shed light on whether reward sensitivity differentially contributes to the development of RJA and IJA. Neuroimaging studies of typically developing children show that IJA may be more strongly linked to brain regions that handle reward processing. However, in children with ASD, an executive function measure thought to reflect reward processing was linked to a combined measure of both IJA and RJA.
This review underscores the significance of both Responding to Joint Attention (RJA) and Initiating Joint Attention (IJA) in the development of Autism Spectrum Disorder (ASD). There is limited evidence suggesting that IJA is more of a core deficit: both RJA and IJA atypicalities are observable in older individuals, and there is restricted evidence indicating that reward processing contributes more to IJA than RJA.
Moreover, both IJA and RJA predict social development, challenging the notion that one is more fundamental than the other. The idea that dyadic abnormalities come before triadic ones is not well supported by strong evidence. This fits more with the Parallel Distributed Processing (PDP) model of joint attention than with the social-cognitive model.
Considering joint attention as a tool for learning from others, changes in joint attention over time may offer valuable insights, as such changes imply the influence of some factors. Thus, assessing joint attention and other developmental domains at various points throughout development is crucial for understanding the mechanisms linking changes in joint attention to other domains.
Joint attention proves to be a powerful predictor across diverse domains in autism and offers a cost-effective and relatively straightforward assessment. Since behavioral tests like joint attention are very good at predicting outcomes, they should be compared to brain imaging and eye-tracking tests made for predicting outcomes in ASD to show how well they work and justify their higher costs, taking into account the common assessment problems that come with eye-tracking or brain imaging.
While joint attention has been discussed as a core difficulty in ASD throughout this review, it's essential to acknowledge that atypicalities in joint attention may arise from strengths. Future research should explore the relationships between covert attention, motor skills, and joint attention in autism. Additionally, there is a need for imaginative exploration of ways that people with autism spectrum disordersmay engage in joint attention but that current measurement systems do not currently capture.