Consider two researchers—Sam and Taylor—who are working on a manuscript with a looming deadline. Both researchers experience a wide range of emotions, from excitement and pride to frustration and disappointment. To work productively towards the deadline, the researchers must manage their reactivity (i.e., how and how strongly a person responds) and continuously engage in emotion regulation (i.e., establishing, maintaining, or altering emotions and their expression; Campos et al., 1989; Rothbart & Bates, 2006). Sam, who tends to be more optimistic while writing, uses various intrapersonal emotion regulation strategies to maintain a positive state, such as reframing the deadline as a challenge. Taylor, on the other hand, who tends to be more pessimistic about writing, looks to Sam for constant reassurance. When Sam uses intrapersonal emotion regulation strategies, Sam is also likely engaging in interpersonal emotion regulation. While Sam’s primary goal might be to regulate Sam’s own emotional state, Sam may also regulate Taylor’s emotional state. Regardless of Sam’s intentions, Sam’s behaviors (e.g., positive emotional expressions) are situational inputs for Taylor, acting as social signals that shape Taylor’s emotional state (Clark & Taraban, 1991; Klinnert et al., 1983). How can we capture the dynamic, continuous, and interlocked nature of affective experiences in real-life dynamics, especially when history (own and shared) has important consequences?

Despite notable progress in understanding the strategies individuals employ in regulating their emotions (Gross, 2015), researchers typically study intrapersonal processes (Campos et al., 2011). This approach does not fully capture the fact that much of affective experiences occur within social contexts involving other people (Frijda & Mesquita, 1994; Keltner & Haidt, 1999; Niedenthal & Brauer, 2012; Parkinson, 1996; van Kleef & Côté, 2022). Speaking to this point, Petrova and Gross (2023) put forth a forward-looking agenda that emphasizes the importance of widening our viewing angle to better capture interpersonal emotion regulation, which includes a “greater theoretical integration of interpersonal and non-interpersonal lines of emotion regulation research.” In our commentary, we discuss how approaches like the Cognitive-Affective Processing System (CAPS) theory (Mischel & Shoda, 1995) provide a useful framework that aligns with and supports this agenda. At their core, such frameworks allow for theoretical integration of the intertwined nature of intrapersonal and interpersonal aspects of emotion and emotion regulation. Here, we provide an overview of key ideas informed by CAPS theory. We then discuss some of the exciting work published in the special double issue in the Future of Affective Science edited by Shiota et al. (2023). We offer forward-thinking suggestions for how such ideas can generate novel future directions.

Informed by cognitive and network models, the CAPS approach conceptualizes a person’s mind as a unique network of cognitive-affective processes. Each person’s Cognitive-Affective Processing System (or “CAPS”) includes encodings and construals (of self, others, situations), affects (feelings and emotions), goals (that drive behavior towards outcomes and emotional states), beliefs (about the world like norms and values), expectancies (about outcomes and efficacy), and competencies and self-regulatory strategies (for achieving desired goals; Mischel, 1973; Mischel & Shoda, 1995, 2008, 2010; Shoda, 1999, 2004; Shoda & Mischel, 1998; Zayas & Shoda, 2009). Each person’s CAPS mediates the effect of the situation on behavior by accounting for how each person encodes, reacts, and ultimately responds to their situation.

Each person’s CAPS network remains largely stable over time and situations, but the particular cognitive-affective processes that are activated at a given moment depend on the situation. What matters is not the objective situation (e.g., the shared task of working on a manuscript) but rather what is psychologically active about the situation (see also Ellsworth & Scherer, 2003; Frijda, 1986; Manstead & Fischer, 2001; Lazarus, 1991). People can differ in their initial evaluation of situational inputs (e.g., Sam reacts less strongly than Taylor to uncertainty and novel situations). But they can also differ in their strategies to regulate emotions (e.g., Sam generally uses reappraisal, whereas Taylor tends to use distraction; John & Gross, 2007). Moreover, the same person can also behave differently—e.g., due to reactivity or preferred strategy—depending on the situation (e.g., Sam is better able to engage in reappraisal when under low stress but fails to do so under high stress; Troy et al., 2013). In this way, CAPS is a person × situation interactionist approach that provides a framework for understanding both between person differences and within individuals.Footnote 1

We focus on the CAPS approach because of its emphasis on psychological mediators that are highly relevant for emotion, emotion regulation, and interpersonal processes (see Ayduk & Mendoza-Denton, 2021; Shoda et al., 2013; Zayas et al., 2021, for applications). We note that other researchers have championed the importance of taking such interactionist approaches. For example, Barrett (2022) argues that contextual factors create a complex web of interactions that give rise to mental events, actions, and psychological meaning that cannot be understood by examining individual components in isolation because they interact non-linearly and produce outcomes probabilistically. Similarly, Doré et al. (2016) have proposed a framework that considers emotion regulation as an interaction of person, situation, and strategy. Doré and colleagues (2016) emphasize that such approaches are necessary to be able to “predict for particular people, in particular situations, which emotion regulation strategies will be most beneficial.”

A key assumption of interactionist approaches is that behavior is intricately connected to the situation that a person is in. But how do we conceptualize a “situation”? By extending the CAPS approach to social interactions, we can conceptualize each interaction partner’s behavior as the situational inputs for the other members present (Zayas et al., 2002). Thus, a dyadic interaction can be seen as the “interlocking” of two individuals’ CAPS. By understanding how each member of the dyad responds to the other’s behaviors, we can illuminate how the dyad operates as an interpersonal emotion regulation system, as well as the intricate connections between intrapersonal and interpersonal emotion regulation (Shoda et al., 2002; Zayas et al., 2002; Zayas et al., 2015). These ideas have several implications for affective science.

First, to the extent that interpersonal patterns are repeated over time, they become internalized as mental representations in a person’s CAPS (Bowlby, 1973; Campos et al., 2018; Holodynski & Kärtner, 2023; Main et al., 1985; Sbarra & Hazan, 2008); thus, interpersonal emotion regulation can become realized intrapersonally (Lee et al., 2024; Zayas et al., in press). Going back to our example of the two researchers, Taylor may develop emotion regulation scripts that take the form of if stressed or upset about writing, then seek support from Sam. Taylor may also develop expectancies; to the extent that such bids for support are effective, Taylor will also develop the expectation that takes the form of if obtain support from Sam, then experience affective relief (see also Waters & Waters, 2006). Eventually, it is possible that simply seeing Sam would provide a sense of calmness and a positive outlook (Selcuk et al., 2012). Moreover, it is possible that Sam will also develop expectations. If Sam’s own emotion regulation attempts positively impact Taylor, then Sam may be more likely to enact reappraisal in the future, resulting in automatic emotion regulation (Mauss et al., 2007).

Second, to the extent that in dyadic interaction, one person’s situation is the other person’s behavior, then a person’s affective experience, including the emotions experienced and emotion regulation strategies used, will depend on who they are interacting with. If Taylor interacts with Sam, who engages in reappraisal, Taylor’s own positive mood and goal-relevant behaviors are maintained. On the other hand, if Taylor interacts with another colleague who also feels daunted by a looming deadline, Taylor’s affective experience and behaviors will differ.

To demonstrate the CAPS approach’s utility, we discuss work from the special double issue and offer suggestions for future directions.

Shore et al. (2023), using a prisoner’s dilemma game in dyads, found that the partner’s regulation of positive expressions and the actor’s perception of this regulation both significantly increased the probability of subsequent actor cooperation, even after controlling for prior defection. These results illustrate the dynamic interplay between situational cues (partner’s positive regulation) and individual perceptions (actor’s perception of the regulation) in shaping cooperative behavior, aligning with the present CAPS framework’s emphasis on the dyadic system (Shoda et al., 2002; Zayas et al., 2002; Zayas et al., 2015). An interesting possibility for future work on interpersonal emotion regulation might be to examine how actors make sense of their partner’s emotion regulation efforts, including their partner’s initial reactivity, and when do they lead to benign inferences or mistrust.

A promising route for future work is to apply a CAPS lens to often-used and emerging tasks in affective science described in the special double issue, such as in psychophysiology (e.g., Qaiser et al., 2023), ambulatory sensing (e.g., Hoemann et al., 2023), and daily diaries (e.g., Tran et al., 2023). For instance, Hoemann et al. (2023) present a biologically triggered, idiographic experience sampling method. Replicating previous research, a key finding is that the same emotion word can correspond to different physiological responses and affect ratings (Hoemann et al., 2020; see also Campos et al., 2004). Their multimodal ambulatory sensing method is particularly suited for if… then… analysis. For example, researchers could also assess the features of psychological “situations” participants find themselves in, alongside assessing the emotions that individuals are experiencing. This includes examining their goals, motivations, and norms, as well as who they are with, during different sampled moments (Tamir & Hu, 2024; Tamir et al., 2020; Vishkin & Tamir, 2023). By analyzing these psychological ingredients, we can more precisely examine how emotional experience is shaped by interpersonal factors. Future work might recruit dyads or groups of friends and use this method to assess emotion contagion, synchrony, and interpersonal emotion regulation.

Similarly, Tran et al. (2023) sampled participants “in the wild” and conducted two daily diary studies. Participants reported having both self-focused (i.e., changing own emotions) and other-focused (i.e., changing other’s emotions) emotion regulation motivations in approximately one in five of the recorded social interactions. Future work could collect more information about the situation, such as who the interaction partner was, what the topic of the interaction was, how long the interaction lasted, how personally significant the topic was, and how well the conversation went. This information could then be used to identify psychological features of “significant social interactions” and how certain psychological features covary with self-focused and other-focused emotion regulation motivations. For example, perhaps people are more likely to report having self-focused and other-focused emotion regulation motivations in private and more intimate settings and topics, or with those who are closer. After identifying potential psychological ingredients, work could then experimentally manipulate situations people are in to obtain causal evidence.

The CAPS approach can also take work that has been primarily focused on intrapersonal processes to an interpersonal context. For example, Becker and Bernecker (2023) highlight the role of hedonic goal pursuit in self-control and self-regulation. Given that self-control varies across situations (e.g., financial, relational; Weber et al., 2002), future work could examine how the strategies people use might differ based on their own reactivity (e.g., towards temptations) and the perceived difficulty of the situation, even if hedonic goals stand in the direct service of long-term goals. The same amount of objective temptation may elicit different strategies depending on the individual’s reactivity and situational context, particularly when considering how these dynamics play out in social contexts.

Taken together, the CAPS approach provides a framework for understanding the dynamic and complex nature of affective experiences. As researchers have long acknowledged, the boundary between intrapersonal and interpersonal emotion regulation is fuzzy (Campos et al., 2004; Petrova & Gross, 2023; Zaki & Williams, 2013). By modeling the mind as a network of cognitions and affects that mediate the impact of situations on behavior, and by considering how in dyadic interactions, two such minds are intertwined, the field will better capture both intrapersonal and interpersonal aspects of emotion and emotion regulation. This approach will refine theoretical models, enabling them to represent the intricate dynamics of everyday affective experiences and inform future research in the field.