Guest Blog Authored by Sho Araiba and Marija Čolić
Sho Araiba, Ph.D., BCBA-D, is a behavior analyst and a filmmaker. He earned his doctorate in psychology (behavior analysis) from the City University of New York. His research interests encompass a wide range of philosophical, basic, and applied topics within behavior analysis. With 15+ years of experience as an ABA therapist, he has served neurodiverse people both in the U.S. and Japan. He also teaches at Leeward Community College, the University of Hawaii. He is also known as Dr. Sho, the creator and host of the Dr. Sho Show on YouTube: https://www.youtube.com/c/DrShoShow.
Dr. Marija Čolić, Ph.D., BCBA-D, LBA, earned her bachelor’s and master’s in psychology and Ph.D. in special education from the University of Belgrade, Serbia. With over 15 years of experience working with autistic individuals, she has been active in the field of behavior analysis since 2014. Currently, Marija teaches at the University of Hawaiʻi at Mānoa on the beautiful island of Oahu. Marija has contributed to the ABA field through multiple articles on family stigma and the cultural aspects of behavior analysis in leading journals. Since May 2020, she has been hosting free webinars for Serbian parents of children with disabilities and professionals on ABA. Her primary research interests include exploring stigma among families of individuals with disabilities and developing ways to support parents. She is also the principal investigator in exploring play among autistic children using a neurodiversity-affirming approach. Marija is an active member of several non-profit organizations, including the Hawai’i Association for Behavior Analysis and Parents for Public Schools Hawaii. Beyond her academic pursuits, Marija is passionate about surfing, the ocean, and marine life.
Have you ever wondered what kind of behaviors are involved in the many different games children play? And equally importantly, how do we, practitioners, support children we serve to play with their peers if that is what they want to do? If a child wants to play tag with peers, you might develop a task analysis. First, find a person who is “it.” Second, run away from “it.” Third, if the “it” tags you, you become “it.” Fourth, you chase your peers. Fifth, you tag one of your peers. Go back to the first step and repeat. However detailed your task analysis may be, it might not be effective in supporting the child in playing with their peers if it does not contain the key behavior repertoires: elementary social behavior. In this blog, we explore a new concept of elementary social behavior in relation to social play and verbal behavior. For the past two years, we, the authors of this blog, have been fortunate to observe and learn about how autistic children play in a group and how we can support and facilitate their social play. We want to share with you what we have learned so far in this blog.
Let Children Lead: Neurodiversity-Affirming Play Matters
For too long, autistic children’s play has been viewed through a deficit lens, focusing on what they “can’t do” instead of celebrating their creativity, curiosity, and strengths (O’Keeffe & McNally, 2025). We believe that it is time to shift this narrative. Autistic play should be about choice, joy, and connection. Thus, when we incorporate social play in our practice, first and foremost, we make sure a child wants to play with their peers. A child needs to feel safe to play with others, to be able to say “no”, and to explore play in ways that feel authentic to them. It is also important to understand that rules are secondary to play: some children love bending or breaking them as part of the fun. Even if a child doesn’t fully understand every rule, they can still participate and create joyful play experiences in their own way. Time, duration, and setting should also honor the child’s preferences. Some children may engage in play for a few minutes, others for longer, and it is always okay for a child to step away to play by themselves. By following what children love to do, we embrace a neurodiversity-affirming approach that honors autonomy, celebrates differences, and ensures that play is always fun and never framed as a task.
In our own social play sessions for autistic children, we also recognize that some kids prefer to play on their own. We provide materials and activities that allow solitary independent play, honoring each child’s comfort and preferences. We learned that when they feel ready or develop an interest in a specific group game, they join in a group independently.
What is Elementary Social Behavior?
Elementary social behavior is a proposal to categorize various social behaviors, including social play behavior, functionally and descriptively, just as Skinner (1957) categorized language functionally as verbal behavior. We define play as an activity that is maintained by social reinforcers (more on this later), and social play as play that involves other peers, unlike solitary play, where a child plays on their own (e.g., Xu, 2008). Our aim is to identify various functional relations between social behavior and social consequences that are important for a person to play with others, and for practitioners to help facilitate their involvement in play.
A word of caution is that this new concept of elementary social behavior is still just a small idea, much like an intellectual exercise Skinner initially engaged in when he first developed the concept of verbal behavior. We have presented preliminary data at various conferences, both in poster form and oral presentations (Araiba, 2025; Čolić, 2025; Čolić et al., 2025; Felipe et al., 2025), but it has yet to accrue empirical support. We would like to know whether this small idea can find a place in your practice.
Social Reinforcers in the Context of Social Play
Social play, such as tag or pretend play, is maintained by social reinforcers. So, we need to first consider the nature of social reinforcers. Our textbook gives examples of social reinforcers such as physical contact, proximity, attention, and praise (Cooper et al., 2020). In verbal behavior, Skinner (1957) also discusses social consequences as generalized conditioned reinforcers. For example, he defined the tact response as an operant response occasioned by a nonverbal discriminative stimulus and maintained by a generalized conditioned reinforcer. Let’s say a child says, “Dog!” upon seeing one (a tact response), and their mother provides attention, “Wow, what a cutie!” (a generalized conditioned reinforcer). In the context of elementary social behavior, we need to pay closer attention to the nature of social reinforcers. For example, the child’s tact response might be uniquely maintained by the mother’s attention and not by other people’s, such as the dog owner, an important difference. To illustrate, Jerome and Sturmey (2013) found that adults with developmental disabilities responded better in the context of completing academic demands with preferred compared to nonpreferred staff members; the effectiveness of social reinforcers partly depended on who delivered them. This preference for attention from preferred individuals is likely even more pronounced in the context of play. To identify social reinforcers for play, we need to know, for example, whose attention functions as a social reinforcer, what actions (e.g., physical interactions, gestures, and verbal responses) function as a social reinforcer, and when and how often social reinforcers should occur to maintain play behavior or to build new play repertoires, among many other factors.
An important parameter of a social reinforcer in social play is when and how often a social reinforcer is provided during play. Let’s look at our task analysis of the tag game. What reinforcers are maintaining participation? The task analysis ends when a child tags another child (and then goes back to the beginning and repeats the steps). If so, tagging a peer should function as a terminal reinforcer and maintain this response chain. If that’s the case, do we, practitioners, reinforce (with a praise, a token, or something else) every time a child tags a peer (or an adult if practicing)? We hope you say, “That doesn’t make sense!” You would point out that the fun part (social reinforcers) of the tag game is not tagging someone, but jostling (the close, physical, and dynamic back-and-forth between the players as they try to avoid being tagged). If that’s the case, social play is maintained by social reinforcers that occur during the activity and not at the end of a response chain.
Play, therefore, is not a response chain, but an activity that is maintained by multiple social reinforcers, which occur many times during play (see Baum, 2002, for the definition of an activity). This brings up the question of whether the task analysis approach to facilitate play behavior is a good idea. Task analysis is good for complex behavior in the form of a chain of responses, and completion of the last step in the chain results in contacting reinforcers (e.g., buying a bottle of water from a vending machine). This characteristic of the task analysis approach does not capture the nature of play, where social reinforcers occur from different peers at any point during play. Furthermore, a task analysis might also accidentally give children the wrong idea of what the play is about. Suppose we tell the child the goal of the tag is to tag someone. This scenario might cause difficulty in play for some autistic children, who prefer to adhere to the rules and have difficulty either breaking or tolerating others breaking the rules (although their peers may actually be engaging in jostling, not breaking the rules).
To summarize, when we facilitate social play with children, we first need to consider the nature of social reinforcers. The effectiveness of social reinforcers depends on many parameters, such as who delivers them, what form of action they take, and when and how often such actions occur during play. Play, in general, is not a response chain. But if play is not a task analysis, how can we support such behavior?
How to Nurture Social Reinforcers
Suppose tag is not a response chain that ends with a terminal reinforcer, but an activity maintained by social reinforcers scattered throughout the interaction. How can we arrange the environment so that the children we serve contact those reinforcers during play?
One idea is to expand a child’s social reinforcer class. In the early stages of development, physical forms of social interaction, such as being held, tickled, tossed onto a bed, kissed, rocked, and engaging in rough-and-tumble play, dominate as social reinforcers in a young child’s life. As the child grows, their social reinforcer class might expand to gestural social reinforcers, such as lightly touching or teasing, pretending to wrestle, laughing, screaming, making faces, and various gestures. Later, vocal interactions such as praising, congratulating, lying, joking, tricking, and commenting join the child’s social reinforcer class. In this sense, jostling in tag is primarily a gestural social reinforcer. If so, we can first introduce games with physical social reinforcers, such as tickling each other and rolling on the mat together, and then we can move on to games with gestural social reinforcers, such as tag and Simon Says. We, practitioners, can also assess children’s social reinforcer class repertoire and see if we can help expand it.
For one, practitioners aim for their own behavior to function as part of the child’s social reinforcer class. Accordingly, we often emphasize establishing rapport with the child by engaging in play with them (e.g., Araiba & Čolić, 2022). By establishing rapport, we hypothesize that our physical, gestural, and vocal behavior enters the child’s social reinforcer class. This might be related to the limited effectiveness of a procedure that pairs our vocal behavior, such as praise, with primary reinforcers such as food, to establish our praise as a conditioned reinforcer (e.g., Dozier et al., 2012). For our vocal behaviors to be a part of a social reinforcer class of a child, we may need not only to pair our vocal behavior with primary, tangible reinforcers, but also to play with the child so that our physical and gestural behaviors join the child’s social reinforcer class first before our vocal behavior to join the class to be an effective social reinforcer.
This is likely true for both neurodivergent and neurotypical children. In our own experience, we are increasingly becoming aware that different types of play come with different social reinforcers. Play such as rolling on the mat, dancing, and passing a ball to each other involves physical social reinforcers, whereas other types of play, such as a telephone game and pretend play, involve vocal social reinforcers. We found it easier to start our social play session with types of play that involve physical social reinforcers, then gradually move to those that use gestural and vocal social reinforcers, for children to play together and play longer. This way, children find each other’s physical, gestural, and vocal behavior as social reinforcers gradually and naturally.
Who is Playing? The In-group/Out-group Behavior
With the analysis of a social reinforcer in mind, let’s now look at a few elementary social behaviors that are a function of social reinforcers. When children play tag, there are many behaviors that we missed in our task analysis. Or rather, our task analysis is based on assumptions that a child has some unspecified behavioral repertoire that is so common-sense that we failed to notice it. For example, let’s say a child plays tag with four other peers in a park. From the get-go, we assume that the child can identify those four peers as members of a play group. When you think of it, it’s a pretty complex discrimination task. The child not only needs to discriminate between each of the four peers but also those four from others in the park, as there are other children (and adults) in the park. On top of this, all of them are moving around constantly (this is not a tabletop match-to-sample task!). For now, let’s call this discriminated operant the in-group/out-group behavior in the context of elementary social behavior. A child discriminates between those who are in-group (a class of individuals who signal the availability of social reinforcers) and those who are out-group (a class of individuals who signal the non-availability of social reinforcers or even the availability of punishers).
The in-group/out-group behavior is not itself a new concept for practitioners. There are instructional procedures that address discrimination of people of different kinds, such as familiar people vs. strangers (especially for safety reasons, e.g., Fisher et al., 2013). We are interested in the development of this in-group/out-group behavior during childhood and whether we can support this development through play. We have observed in our program that children initially try to involve their parents in their play. For example, if we engage in a pretend restaurant play, the children would go to their mothers to serve some food, even though the parents are sitting away from the children and are not “in” the game. This behavior later decreased in frequency as the sessions progressed over time, which indicates that children learned to discriminate between those who were playing and those who were not, as children became more familiar with each other and received stronger social reinforcers from their peers compared to their parents.
Where is the Play? The Territorial Behavior
Another unspecified aspect of the tag is the notion of the territory in which play takes place. These five children who play tag are in the park, and they will not leave the park during the play. The task analysis says “run away from it,” but where and how far should the child go? For the tag to be maintained, all five children need not only to be in one area (the park) but also in physical proximity with each other (so that there is always a chance to tag and be tagged). If the child leaves the park or is too far from their peers, the game of tag breaks down because jostling is not possible. We call this elementary social behavior the territorial behavior, in which a child’s physical proximal distance from their peers is a function of the value of the social reinforcers (e.g., as the distance between the children increases, the value of social reinforcers decreases). Again, we, practitioners, are not new to this behavior. Researchers have worked on many similar behaviors, such as addressing elopement (e.g., Piazza et al., 1997) and staying close to a caregiver in public (e.g., Miltenberger & Novotny, 2022). In our play sessions, we also observed children gradually learn to play in a designated area over time.
Who is “It”? The Role-Taking Behavior
At this point, you must be screaming, “The child needs to learn what the ‘it’ is!” Play of tag relies heavily on this symbolic behavior of assigning someone “it.” Now, what is this “it?” Is this a stimulus just like the golden arch is the symbol of McDonald’s, or a word that is a symbol of an object, as many stimulus equivalence researchers speculate (e.g., Horne & Lowe, 1996)?
Developmental psychologists have a long history of identifying and categorizing the play behavior of young children. Lifter and her colleagues (e.g., Lifter et al., 2011) and the pivotal response training group (e.g., Stahmer, 1995), in particular, have worked on symbolic play and pretend play, and developed many interventions that facilitate these types of play among autistic children. In essence, symbolic play requires tact extension such that a child tacts a carrot as a car and moves it as if it is a toy car. In pretend play, such tact extension becomes more complicated as a child assigns certain characteristics to a doll, engages in intraverbals with a doll, and so on (e.g., the doll is a guest at a tea party).
What is going on here? A child is assigning a role to objects and people. A role is a stimulus function of a person or an object whose behavior is specified by a certain class of behavior. For example, the role of the mother consists of a class of behaviors such as feeding, bathing, dressing, giving attention, and so on. As the child’s mode of play shifts, this role becomes more and more complicated from assigning a role of a car to a carrot (symbolic play) to assigning a role of a guest to a doll in a tea party (pretend play) to assigning a role of “it” to a peer in the tag (cooperative play), to assigning a role of a pitcher to a teammate in baseball (rule-based games). Thus, the “it” in the tag can be conceptualized as a type of role. In elementary social behavior, we call this role-play behavior. There are two parts to the role-play behavior. One is the role-play speaker behavior, where a child takes up a role and engages in a role-based class of responses (a child becomes “it” and starts chasing peers). The other is the role-play listener behavior, where a child acts in response to the role-play speaker behavior (a child runs away from “it”). Social reinforcers likely occur when both the speaker and the listener act according to their roles.
Many More Elementary Social Behaviors
By now, you can see what we are up to with elementary social behavior. We are realizing that there are many elementary social behaviors that children engage in to play a game. We are running out of space to share some more that we discovered, but we hope to share them somewhere else in the near future. For now, let’s consider how the tag ends. As we pointed out earlier, tag is not like sports or board games that have a discrete beginning and ending (it is not a response chain). Tag is an activity just like rough-and-tumble play, so the end of tag is often due to satiation of social reinforcers. When one or more in-group members stop engaging in running or chasing, that indicates that they are likely satiated with social reinforcers or that they are physically exhausted, and, consequently, the tag game ends. Thus, a child needs to discriminate their peer’s motivating state as well as their own (Lewon’s basic research shows the discriminative properties of motivating and emotional states: see Lewon & Heyes, 2014; Stites et al., 2022). In elementary social behavior, we call this the state-discrimination behavior, a discriminated operant as a function of the changing availability and non-availability of social reinforcers of the in-group members. Children benefit from this discrimination because over-satiation will lead to the devaluation of a social reinforcer for peers in the future (in other words, peers might not be motivated to play in the near future as they are over-satiated).
Conclusion
In conclusion, in this blog, we shared a small new idea that we call elementary social behavior. Elementary social behavior is a functional and descriptive taxonomy of social behavior with a special focus on social play. The primary reason why we are working to identify elementary social behavior is to find an effective and efficient way to facilitate social play among children. Although there are no definitive answers yet, our research suggests that a gradual introduction to different types of social play with different types of social reinforcers would benefit autistic children to play more frequently and longer with their peers. As we disclosed at the beginning of this blog, what we wrote here is just an idea that has not yet been scientifically validated. But we hope this will inspire you to look at your own practice of social play and social skills group programs more closely and to help facilitate children’s wish to play with their peers. And stay tuned for updates on our project!
References
Araiba, S. (2025, June 21-23). Behavior Analysis of Social Play -A Primer. Connecting ABA World Autism Summit. Hakodate, Japan.
Araiba, S., & Čolić, M. (2023). Preliminary practice recommendations for telehealth direct applied behavior analysis services with children with autism. Journal of Behavioral Education, 32(4), 768-802. https://doi.org/10.1007/s10864-022-09473-6
Baum, W. M. (2002). From molecular to molar: A paradigm shift in behavior analysis. Journal of the experimental analysis of behavior, 78(1), 95-116. https://doi.org/10.1901/jeab.2002.78-95
Čolić, M. (2025, September 4-5). A play-based approach to increasing play and social skills in autistic children. The 20th Annual Hawai’i Association for Behavior Analysis Convention, Honolulu, HI, USA.
Čolić, M., Araiba, S., & Felipe, J. (2025, February 4). Playfully proficient: Enhancing social skill acquisition in young children with autism through play-based intervention [Poster presentation]. Research Day at the State Capitol Building, Honolulu, HI.
Cooper, J. O., Heron, T. E., & Heward, W. L. (2020). Applied behavior analysis. Pearson UK.
Dozier, C. L., Iwata, B. A., Thomason‐Sassi, J., Worsdell, A. S., & Wilson, D. M. (2012). A comparison of two pairing procedures to establish praise as a reinforcer. Journal of Applied Behavior Analysis, 45(4), 721-735. https://doi.org/10.1901/jaba.2012.45-721
Felipe, J., Čolić, M., & Araiba, S. (2025, September 4–5). Supporting autistic children in social play through fun games [Poster presentation]. The 20th Annual Hawai‘i Association for Behavior Analysis Convention, Honolulu, HI, United States.
Fisher, M. H., Burke, M. M., & Griffin, M. M. (2013). Teaching young adults with disabilities to respond appropriately to lures from strangers. Journal of Applied Behavior Analysis, 46(2), 528-533.https://doi.org/10.1002/jaba.32
Horne, P. J., & Lowe, C. F. (1996). On the origins of naming and other symbolic behavior. Journal of the Experimental Analysis of Behavior, 65(1), 185–241. https://doi.org/10.1901/jeab.1996.65-185
Jerome, J., & Sturmey, P. (2008). Reinforcing efficacy of interactions with preferred and nonpreferred staff under progressive‐ratio schedules. Journal of Applied Behavior Analysis, 41(2), 221-225. https://doi.org/10.1901/jaba.2008.41-221
Lewon, M., & Hayes, L. J. (2014). Toward an analysis of emotions as products of motivating operations. The Psychological Record, 64(4), 813-825. https://doi.org/10.1007/s40732-014-0046-7
Lifter, K., Foster-Sanda, S., Arzamarski, C., Briesch, J., & McClure, E. (2011). Overview of play: Its uses and importance in early intervention/early childhood special education. Infants & Young Children, 24(3), 225-245. https://doi.org/10.1097/IYC.0b013e31821e995c
Miltenberger, R. G., & Novotny, M. A. (2022). Teaching safety skills to individuals with developmental disabilities. Advances in Neurodevelopmental Disorders, 6(3), 270-279. https://doi.org/10.1007/s41252-022-00248-8
O’Keeffe, C., & McNally, S. (2025). ‘Like it’s making my heart run’: A strengths-based understanding of the play of autistic children. Autism, 29(6), 1469-1482. https://doi.org/10.1177/13623613251315985
Piazza, C. C., Hanley, G. P., Bowman, L. G., Ruyter, J. M., Lindauer, S. E., & Saiontz, D. M. (1997). Functional analysis and treatment of elopement. Journal of Applied Behavior Analysis, 30(4), 653-672. https://doi.org/10.1901/jaba.1997.30-653
Skinner, B. F. (1957). Verbal behavior. Appleton-Century-Crofts.
Stahmer, A. C. (1995). Teaching symbolic play skills to children with autism using pivotal response training. Journal of Autism and Developmental Disorders, 25(2), 123-141. https://doi.org/10.1007/BF02178500
Stites, M., Lewon, M., Peters, C. M., & Hayes, L. J. (2022). Motivating operations as contexts for operant discrimination training and testing. Behavioural Processes, 203, 104779. https://doi.org/10.1016/j.beproc.2022.104779
Xu, Y. (2008). Children’s social play sequence: Parten’s classic theory revisited. Early Child Development and Care, 180(4), 489–498. https://doi.org/10.1080/03004430802090430