Social Network Types in Autistic Adults and Its Associations with Mastery, Quality of Life, and Autism Characteristics

In the general population, experiencing social support is positively related to mental health (e.g., Gariépy et al., 2016; Santini et al., 2015). Research in autistic adults on social support and social networks is relatively limited, especially in middle-aged and older autistic adults (Charlton et al., 2022). However, we know that social support is a consistent predictor of quality of life and well-being in autistic individuals (Bailey et al., 2020; Bishop-Fitzpatrick et al., 2018; Charlton et al., 2022; Deserno et al., 2017; Leader et al., 2021; Tobin et al., 2014). In addition, concepts close to social support seem to play a role in mental health in autistic adults. For example, higher levels of social isolation are related to higher levels of social anxiety and depression (Schiltz et al., 2021), and increased loneliness is related to higher levels of psychological distress (Moseley et al., 2021). Also, vice versa, satisfaction with received social support is related to decreased depression symptoms (Gotham et al., 2014). Mental health problems are known to have an elevated prevalence among autistic individuals compared to the general population (Hollocks et al., 2019; Hudson et al., 2019; Lai et al., 2019), with lifetime prevalences estimated between 14.4 to 37% for depression (Hollocks et al., 2019; Hudson et al., 2019) and 42% for anxiety disorders (Hollocks et al., 2019). In this study, we further expand the knowledge on social networks of autistic adults, by exploring the typology of social networks and its associations with quality of life, autism characteristics, and sense of mastery.

Mastery, described as the extent to which individuals feel in control over their lives (Pearlin et al., 1981), is thought to be positively influenced by social support (Thoits, 2011). Also, vice-versa, individuals with a high level of mastery might be more able to form a supportive social network (Thoits, 2011). For example, they might be more likely to choose the social activities that they actually experience as favorable or supportive. Indeed, autistic women stated that another person’s belief in their abilities contributed to the belief in their own abilities (Webster & Garvis, 2017). Research indicates that autistic adults report lower levels of mastery compared to non-autistic adults (van Heijst et al., 2020). Mastery was also identified as a key variable to distinguish two validated subgroups in autistic adults (Radhoe et al., 2023b). Membership of these subgroups was predictive of the quality of life experienced by autistic adults up to five years in the future (Radhoe et al., 2024). Besides these findings, little research has been done on mastery within autism populations, although there is research on related concepts that indicate feeling in control over one’s life outcomes might be related to quality of life in autistic adults. For example, a study in autistic adults showed that higher levels of psychological empowerment, here operationalized as one’s perception of control, efficacy, and outcome expectations, was related to increased quality of life (White et al., 2018). Also, autism traits seem to affect quality of life indirectly through the mediation of feelings of competence, with the latter conceptualized here as feeling one has skills to make effective choices that lead to achieving expected and valued outcomes (Andrews et al., 2023). Thus, increased understanding of feelings of mastery and its relationship to social support in autistic adults could provide a new entry point to improve quality of life.

Autistic individuals seem more vulnerable in their social support networks compared to non-autistic individuals as they experience lower social support levels (Alvarez‐Fernandez et al., 2017; Bishop-Fitzpatrick et al., 2018), their network is often smaller (van Asselt-Goverts et al., 2015), and they are more socially isolated (Stewart et al., 2024). However, research on this topic also suggests that there is a large heterogeneity in social network aspects within the group of autistic individuals, which is consistent with the notion of large heterogeneity between individuals on the autism spectrum in general (Agelink van Rentergem et al., 2021; Lombardo et al., 2019). For example, social network size and density differed substantially among autistic university students (Lei et al., 2019), with some students having small and dense networks, and others having large networks with low density. Further, social interaction is experienced very differently among autistic adults, where some described this as anxiety-provoking or exhausting, but others as mainly enjoyable (Ee et al., 2019). Struggle to form social connections is frequently reported in autistic students, but the type of reported difficulties ranged rather broadly (Bailey et al., 2020). Some reported a desire for a romantic partner, but apart from that they were relatively satisfied with their friendships, while others felt that forming any meaningful relationship was not possible for them (Bailey et al., 2020). Research on subtypes of social support networks within autistic adults could potentially clarify this heterogeneity in social support experiences.

Studies on individual social network aspects (e.g., network size or specific types of social support such as emotional or practical support) and its associations with external variables such as mastery and quality of life provide us with important information on the role of social networks in the lives of autistic people and on possible targets to improve quality of life. However, adopting a multivariate approach using subtypes instead of individual social network aspects to investigate associations with external variables reduces the complexity of (1) intercorrelation between aspects, and (2) combined effects of multiple aspects on external variables (Radhoe et al., 2021). Studies in other populations, such as individuals with an intellectual disability (Tournier et al., 2021) and individuals with long-lasting mental health conditions such as schizophrenia or bipolar disorder (Sweet et al., 2018), have been able to distinguish typologies of social networks with meaningful associations to well-being. Considering the importance of social support for quality of life in autistic people, interventions aimed at improving social support or social networks in autistic adults have been recommended (Bishop-Fitzpatrick et al., 2018; Moseley et al., 2021).

Knowledge on (1) what network types exist within this group and (2) how specific network types relate to relevant external variables might guide clinicians for which individuals such social network interventions apply most. Also, by relating these network types to feelings of mastery and quality of life, we can better understand the heterogeneous experiences of autistic individuals with social support networks, and better identify what aspects are important for whom. Therefore, the current study explored if we could distinguish different types of social networks in autistic adults, based on both functional (e.g., perceived social support) and structural (e.g., network size) aspects. Additionally, we explored associations between the identified social network types and three external variables: quality of life, autism characteristics, and mastery.

Autistic individuals show large heterogeneity regarding many aspects, including in their social experiences. In this study, we explored whether we could identify meaningful subtypes in social support networks in autistic adults that relate to psychological factors. Three clusters emerged, which can be described as clusters with participants who had: 1) at least two close persons, sometimes including a romantic partner; 2) only a romantic partner as a close person; and 3) no close persons. Here we will summarize and further expand on the results of this study.

First of all, it is important to note that the identified social support network clusters did not differ in autism characteristics or quality of life in terms of physical health, psychological health, and environmental factors. Also, sense of mastery was not related to one of the identified social support network clusters, contrary to what has been theorized before on the relation between mastery and social support (Thoits, 2011). This indicates that clusters of social support networks are not predictive of the external variables examined. This raises the question of whether such associations between social support network types and mastery and quality of life aspects are absent in autistic individuals, or whether they might emerge if other social support characteristics were used in the cluster analysis. As this is a first exploration of social support network clusters in autism, further research is needed to draw firm conclusions about predictive value. Future studies should still consider including mastery as external predictor, given its established links to mental health (e.g., Burns et al., 2011; Dalgard et al., 2007) and quality of life (e.g., Raaijmakers et al., 2014) in other populations, and the limited number of studies that investigated mastery in autistic individuals (e.g., Radhoe et al., 2023b; van Heijst et al., 2020).

Second, it is not unexpected that a small cluster of participants without any close persons emerged (i.e., Cluster 3), given that autistic individuals described both difficulties in establishing and maintaining social relationships, and desires to increase and deepen such contacts (e.g., Ee et al., 2019; Müller et al., 2008; van den Heuvel et al., 2023). A recent study (Stewart et al., 2024) also identified a small group of middle- and older-aged autistic adults who had no family or friends with whom to discuss private matters. The desire for social contact can be ambivalent for autistic people because social situations can be energy demanding and stressful (Kanfiszer et al., 2017; Kock et al., 2019), which can even be enhanced by previous negative social experiences such as being bullied (Ee et al., 2019; Kanfiszer et al., 2017; Sedgewick et al., 2019). The reasons for the absence of close relationships in Cluster 3 are speculative, but we want to emphasize that individuals in this cluster reported lower quality of life about social relationships. So, participants in Cluster 3 seemed to be unhappy with their small social network. This is consistent with research showing that autistic adults experience high levels of loneliness and long for connection and belonging (Grace et al., 2022). Hence, individuals in this Cluster 3 appear particularly crucial to involve in social network interventions to see what they need to satisfy their social needs.

Third, no difference was found in satisfaction with social relationships between those in Cluster 1 (i.e., multiple close persons, sometimes including a partner) and 2 (i.e., only a partner). Apparently, this quality of life domain does not depend on the number of close persons, as long as there is such a close person with whom one has a sense of belonging or support. This is in line with previous findings, for example that satisfaction with one’s network has a more direct effect on well-being than the number of social contacts an autistic individual has (Deserno et al., 2017). Also, a qualitative study found that having one close relationship seemed to work as a buffer for social isolation in autistic older adults (Hickey et al., 2018). Previous research described the feeling of belonging as important for well-being in autistic people, because it makes a person feel understood and accepted (Milton & Sims, 2016). Also, a review of lived experiences of autistic girls and women pointed at the importance of social factors, such as social connectedness, as determinants for well-being and mental health (O’Connor et al., 2024). Inclusivity, support, and acceptance within friendships and relationships were described as important aspects (O’Connor et al., 2024). Feelings of social isolation and loneliness are more common in autistic individuals compared to non-autistic individuals (Grace et al., 2022). This may stem from unmet desires for social contacts (Elmose, 2020), but can also be experienced as a form of isolation within oneself despite having good relationships with others (Hickey et al., 2018). Importantly, lack of autism understanding, and experiences of “othering” as described by autistic people (Elmose, 2020; Milton & Sims, 2016) hinder social connectedness and must continue to be addressed in society to improve quality of life of autistic individuals.

Although participants from the first and second cluster did not differ in their quality of life scores regarding social relationships, one could argue that having only one close person (i.e., such as a romantic partner for those in Cluster 2) can make a person more vulnerable. If circumstances change, for example if that one close person develops problems such as mental or physical health problems, it can become challenging to continue providing social support during this period. Also, some autistic women stayed in negative romantic relationships for fear of not finding another partner easily or because their partner was also their access to their partner’s social circle (Sedgewick et al., 2019). In such situations, it might be more difficult to leave a harmful relationship for autistic adults, which could be concerning if this relationship is the only close person in someone’s network. Also, autistic adults who experienced extreme negative experiences within intimate relationships, such as violence or sexual abuse, reported that having friendships could have served as a protective factor (Douglas & Sedgewick, 2024). Thus, although participants in the second cluster had higher quality of life scores than people in Cluster 3, it may still be in their interest to explore whether their circle of close persons could be broadened.

The current study has some strengths and limitations. A strength of this study is that it focused on middle-aged and older autistic adults, which is an understudied group. Moreover, the use of a multivariate approach contributes to the understanding of social support in autistic adults of this age group. There are also some limitations to keep in mind when interpreting the results of this study. First, the current findings might not be generalizable to all autistic individuals due to characteristics of the sample as 1) participants were often diagnosed with autism later in life (M = 46.1; SD = 13.2); 2) participants were highly educated and people with a co-occurring intellectual disability were not included; and 3) the sample was not ethnically diverse (i.e., mainly white). However, the participant group might still be representative of many autistic individuals in mental healthcare, as a substantial proportion of participants was recruited from mental healthcare institutions and participants reported high levels of psychological difficulties (see Radhoe et al., 2023b). Still, replication in other, more diverse, autism samples is needed. It would also be insightful to further explore potential gender differences, considering the distinct male/female ratio in Cluster 2 compared to the other clusters, as well as previously documented differences in friendship experiences between autistic girls and boys (Sedgewick et al., 2016). In addition, future research could test temporal stability and predictive validity of the observed clusters (Agelink van Rentergem et al., 2021) to further finetune knowledge on who would benefit most from social network interventions. A second limitation is that support from groups of people, which can be conceptualized as affiliation networks (e.g., Breiger, 1974), were not taken into account. People can feel a connection to a group of people, without having a special or close relationship with one person within this group. For example, attending autism group meetings after a late autism diagnosis offered a sense of shared experience and understanding (Hickey et al., 2018), so social support from group membership seems important to include in future research. Also, previous research showed that autistic individuals viewed professionals as an important source of social support within their social network (van Asselt-Goverts et al., 2015). Thus, in future research, including the option to refer to both one’s affiliation networks as well as professionals as close persons would help to better understand and support the social support needs of autistic adults. A third potential limitation is that the questionnaires used were not designed or tested for autistic individuals specifically, although our autistic think tank was involved in the instructions of each of the measures used. Measures such as the WHO-QOL might miss aspects of quality of life important to autistic individuals (McConachie et al., 2020; Saez-Suanes & Alvarez-Couto, 2021) but so far this has only been tested in the UK setting and it is not clear whether cultural differences impact which additional questions needs to be added. Similarly, the CPQ may not align well with autistic people’s perceptions on social support aspects, and, for example, when a network member is seen as a close person.

Concluding, this study showed that the absence or presence of close persons can have a significant impact on the quality of life on social relationships in autistic adults. This highlights the importance of addressing social support and relationship satisfaction in interventions aimed at improving the quality of life for autistic adults, which has been put forward before (e.g., Moseley et al., 2021).