The Limits of Simulation for Understanding Mental Illness: Defending a Steinian Theory of Empathy

Abstract

I defend Edith Stein's theory of empathy as an alternative to simulation theories of empathy. Simulation theories of empathy involve using one's own cognitive resources to replicate the mental states of others by imagining being in their situation. I argue that this understanding of empathy is problematic within the context of mental healthcare because it can lead to the co-opting and assimilation of another person's experiences. In response, I maintain that Stein's theory is preferable because it involves appreciating others’ experiences as it is for them, and this alternative account of empathy avoids the assimilation of the experiences of others.

Résumé

Je défends que la théorie de l'empathie d'Edith Stein représente une solution de remplacement aux théories de l'empathie fondées sur la simulation. Les théories simulationnistes impliquent l'utilisation de ses propres ressources cognitives pour reproduire les états mentaux des autres en imaginant être dans leur situation. Je soutiens que cette compréhension de l'empathie est problématique dans le contexte des soins de santé mentale, car elle peut conduire à la cooptation et à l'assimilation des expériences d'une autre personne. En réponse, je soutiens que la théorie de Stein est préférable parce qu'elle implique l'appréciation des expériences de l'autre telles qu'elles sont pour lui, et que cette explication alternative de l'empathie évite l'assimilation des expériences des autres.

1. Introduction

In this article, I explore the advantages of Edith Stein's phenomenological theory of empathy, as an alternative to simulation theories of empathy, for its practical applications within the context of mental healthcare. Simulation theories of empathy involve using one's own cognitive resources to replicate and simulate the experiences and mental states of others by imagining being in their situation. Although simulation theories of empathy should presumably allow us to better understand the experiences of other people in a direct manner, I argue that understanding empathy in this way is problematic primarily because it prioritizes the experiences of the empathizer rather than the experiences of the other person. As an alternative to simulation theories of empathy — within the specific context of mental healthcare — I maintain that Stein's phenomenological theory of empathy offers a solution because it recognizes that empathy involves appreciating others’ experiences as it is for them and, as a result, her conception of empathy allows for interpersonal engagement to occur while avoiding the assimilation of the experiences of others.

2. What Is Empathy?

If you were to ask someone, “What is empathy?,” the likely response you might hear is that empathy involves “putting yourself in the shoes of someone else.” Empathy is typically viewed as the ability to identify with the experiences of others, and it indicates an opportunity for engaging with another person on an emotional and cognitive level. As Birgit Derntl and Christina Regenbogen (2014) note, most models of empathy contain three core components: (1) the ability to recognize emotions in oneself and others, (2) the ability to share the emotional states of others and respond to these affective states in others, and (3) a perspective-taking ability that allows the empathizer to imagine the perspective of others to understand them (Derntl & Regenbogen, 2014, p. 70). These core features are exemplified through simulation theories of empathy.

3. Simulation Theories of Empathy

Simulation theories of empathy — sometimes called “reconstructive” or “re-enactive” empathy — involve a complex process of using one's cognitive resources to replicate the experiences of another person within oneself in order to facilitate understanding with the other person. This replication and simulation of the other person's mental states in our imagination then allows us to gain a deeper understanding of her behaviour (Meneses & Larkin, 2012, p. 156; Ratcliffe, 2012, pp. 474–475). Given our imaginative capacities, our cognitive systems allow us to imagine things that extend beyond ourselves. Since humans have minds that function psychologically in a similar manner that allow us to perform these imaginative feats, the underlying assumption of simulation theories is that we can use our own minds to understand other minds by imagining the world from their point of view (Harris, 2000; Heal, 1995, p. 49; Stueber, 2006, pp. 115–116). Simulation theories are attractive because they offer an objective basis for understanding the minds of others. It is important to clarify that simulation theories are not precisely theories of empathy. Simulation theories are primarily designed to make sense of the experiences of others to predict their behaviour in certain situations. But since they are designed to be neutral, evidence-based, and informed by principles of rationality, approaches like simulation theories are deemed valuable as a means for understanding others because they can serve as a theoretical foundation for gaining future knowledge.

One model of simulation, proposed by Alvin Goldman (2006), is a form of analogical inference comprised of three phases: the matching phase, the simulation phase, and the attribution phase. During the matching phase, the “simulator” introduces “pretend-beliefs” and “pretend-desires” derived from themselves in her own mind to initially match the perceived intentional state of the “target” of simulation. In the simulation phase, the simulator uses her own cognitive resources to process these pretend-beliefs in a way that mirrors the internal cognitive processing system of the other person. Finally, in the attribution phase, the simulator projects the presence of these mental states onto the target (Stueber, 2006, p. 120). At the end of this three-stage process, the simulator should have a fairly good sense of what the other person is experiencing, based on how the simulator would respond or feel in response to similar inputs.

For example, suppose a therapist is trying to understand a person's description of living with chronic depression. During the matching phase, the simulator (the therapist) would first introduce pretend-beliefs about feeling lonely and desolate to match the perceived mental state of the target (the person with chronic depression). Now that the therapist is holding these beliefs of what it is like to be depressed, he would then use his own cognitive resources to see what it is like to be living in this state of mind and what can be learned from this mental exercise (the simulation phase). The therapist might say to himself, “If I were chronically depressed, I would feel hopeless and aimless and I would be looking for ways to ease my pain.” After going through this process of simulating what it is like to be chronically depressed, the therapist then makes an analogical inference that the other person would also be feeling this way (the attribution phase). This gained insight can then help the therapist to offer ways for the patient to manage the symptoms of her depression, or perhaps the therapist might use this information to find alternative treatment strategies to help the patient cope with the feelings of depression.

The strength of Goldman's theory of simulation in a therapeutic context is that it offers caregivers and therapists a way to engage directly with their care recipients in order to get a better sense of what they are experiencing. Moreover, Goldman's approach also bears similarities to our common sense and intuitive conception of empathy as “putting yourself in someone else's shoes.” As a result, Goldman's theory of simulation is easily graspable and often reflects how practices of empathy typically occur in daily life. Moreover, there are several other advantages to simulation that can prove beneficial in the context of mental healthcare.

4. The Benefits of Simulation Theories of Empathy in Therapeutic Contexts

One benefit of simulation theories is that they recognize that people share similar mind states, and this awareness can bridge the epistemic gap between persons. This can be especially helpful when it comes to empathizing with the experiences of persons diagnosed with mental illness for the purposes of fostering connections between caregivers and care recipients. For example, schizophrenia has historically been viewed as a disorder that defies empathic understanding, and it remains one of the most stigmatized and misunderstood mental illnesses. The degree of disorganized thoughts associated with schizophrenia may be so strong that it is incomprehensible for the person experiencing it and this can impair effective communication with others. Furthermore, the presence of a fragmented sense of self reinforces the notion that it is an illness that cannot be understood from an outsider's perspective. But as the cognitive sciences continue to develop and offer further information regarding the brain and how the presence of mental illness influences someone's experiences of the world, the benefit of simulation is it allows for the creation of authentic models of the mind that can provide accurate depictions of what it means to experience certain mental phenomena, including schizophrenia and other mental illnesses.

Another therapeutic benefit of simulation theories is that they can help therapists better diagnose and make sense of their patients’ symptoms. The current shift in psychiatry towards adopting more neuroscientific approaches towards understanding mental disorders with improved accuracy can prove beneficial for improving relationships between caregivers and care recipients. For example, the Research Domain Criteria (RDoc) offers a neuroscientific exploration into mental disorders and seeks to provide a biological reductionist account of mental disorders that makes them easier to identify, diagnose, and potentially treat. But if therapists can also recreate the experiences of patients directly, via the cognitive act of simulation, it might make it easier for them to connect with their patients and to offer insight into their conditions that may not be currently possible under existing diagnostic frameworks.

A third benefit of simulation theories is that “simulations,” as a practice of understanding others, can serve as an invaluable training tool for therapists to help decrease the divide between themselves and those under their care and to help form more positive relations between both parties based on mutual understanding. If it is possible to replicate the experiences of others, and if we can predict their behaviour in response to these stimuli with greater accuracy, then the possibility of utilizing simulation theories to create tailored treatments to meet their specific needs can become a reality. As a result, not only would the quality of care received by persons with mental illness be improved, but this increased understanding may also help to minimize the stigma surrounding mental illness.

5. Objections to Simulation Theories of Empathy

Although there are instances where simulation is unsuccessful, Goldman argues that simulation is the “fundamental method” used for arriving at “mental ascriptions of others” (Goldman, 1995, p. 83). Simulation can often produce “close facsimiles of naturally-generated states” and, as a result, simulation can be used to make sense of the mental states experienced by others (Michlmayr, 2002, p. 26). However, the process of simulation “does not involve the very same states in the attributor as those undergone by the target” (Michlmayr, 2002, p. 25). Thus, while they are not a perfect one-to-one recreation of another person's experiences, the results achieved through this process are sufficiently similar that they can be used to get a better understanding of others and what they are experiencing.

While the aim of simulation theories is to provide an objective and empirical method of understanding the experiences of others, one of the drawbacks of this approach is that it presupposes an ideal knower who serves as the foundation for simulations to occur. When inferring actions and behaviours onto others based primarily on one's own cognitive framework, the simulating agent's cognitive functioning is taken to be “normal,” “standard,” or “neurotypical” and perhaps the presence of mental illness is not factored into the act of simulation. And since the simulating agent makes inferences about others based on his own experiences, this neurotypical agent is unable to understand what someone with schizophrenia is experiencing precisely because the notion of “mental illness” remains outside of the scope of the simulation parameters.

Goldman's theory of analogical inference simulation is problematic in a therapeutic context because the therapist is using himself as the standard model for interpreting his care recipient's experiences. According to Lorraine Code (1995), if caregivers attempt to empathize with care recipients by using themselves as models for interpreting others, it reinforces an epistemic authority that is external from the lived experience of the person diagnosed with the medical condition. Similarly, Peter Goldie maintains that an approach to empathy that uses oneself as a model for others is problematic because it essentially “usurps the agent's own first-personal stance” towards what that person is thinking and feeling, which has the potential of undermining the other person's agency and replacing it with one's own (Goldie, 2011, p. 303). By using ourselves as the standard for understanding others, simulation can further contribute to this demarcation between “neurotypical” and those who are labelled “neurodivergent.”

The purpose of transposing oneself into the situation of the other via simulation can be viewed as a way of diminishing or eliminating contextual differences that exist between individuals, including cultural, historical, and social factors. On face value, this approach seems promising for applications in a therapeutic context because it should allow the therapist to experience a situation as close as possible as the person diagnosed with schizophrenia experiences it. And it should allow the therapist to engage with his patient's phenomenological experiences as neutrally as possible in order to understand it without injecting any personal bias into that encounter that would distort or influence the way the experience is understood (Kögler & Stueber, 2000, p. 23).

But by presuming that we can understand the experiences of the other person to a high degree, simulation theories can be unreceptive to interpersonal differences and may minimize existential differences which exist between people by attempting to experience what the other person is experiencing in the same way (Ratcliffe, 2015, pp. 230–231). While the motivation behind attempting to actualize this impartial and neutral approach towards understanding others is well intended, this striving towards removing cultural, historical, or social differences is problematic precisely because it can result in further instances of epistemic silencing of the perspectives and situated knowledge of traditionally marginalized groups. Thus, while the intention of seeking an objective and neutral standard is admirable, the negative consequence is that, if left unchecked, simulation theories can reinforce certain epistemic modes of knowing as ideal, and they can lead to certain groups of people falling between the cracks since their ways of experiencing the world do not fit these typical epistemological models.

Part of the reason that we can never fully know another person is because a person is “intrinsically unknowable in her entirety” (Ratcliffe, 2015, p. 247, emphasis added). Since the empathizer may impose her own attitudes and beliefs onto the experiences of the other, the most significant objection against simulation theories of empathy is that they can lead to an assimilation of the other instead of a simulation of the other. On that note, Matthew Ratcliffe's analysis raises key insights as to why simulation is insufficient to understand the experiences of others:

Simulation can contribute to a sense of what another person might be experiencing, but empathy demands restraint …. To engage with [another's] experience, an attitude involving openness, curiosity, and reciprocity is needed. Imposing one's own experience on someone … without listening, without being open to alternatives, is a failure of empathy. First-person experience thus informs empathy, rather than serving as a substitute for it. (Ratcliffe, 2015, pp. 245–246, emphasis added)

Simulation theories are problematic because they can potentially result in empathizers co-opting others’ experiences and substituting their own. Overcoming this difficulty requires a theory of empathy that allows for interpersonal engagement to occur yet preserves a distinction between the self and the other. The theory of empathy developed by Stein (1964) is a viable alternative specifically for its benefits in therapeutic contexts.

6. Articulating Stein's Theory of Empathy as an Alternative to Simulation Theories

Before outlining the core features of Stein's theory of empathy, it is helpful to clarify the distinction she makes between primordial (original or direct) experiences and non-primordial (non-original or indirect) experiences. This distinction is significant for differentiating Stein's theory of empathy from alternative approaches, such as simulation theories, precisely because these alternative theories strive to recreate (at least to some degree) the primordiality of another person's experiences within her.

Primordial experiences are any phenomena we perceive firsthand and are given to us fully in our perceptual awareness (OPE §6; Dullstein, 2013, p. 343).¹ For instance, if I am working in my office on a hot summer day and I feel the cool breeze from my air conditioner, the cooling sensation is given to me primordially because I am experiencing it directly. But there are also instances, such as memories or expectations, that are non-primordially given to us (OPE §6). For example, I can remember a joyful experience I once had, such as being accepted into a PhD program, and I can anticipate the joy I will experience once I defend my dissertation. Although both events are experiences I have had in the past or will have in the future, neither of these moments are experienced primordially because their intentional object (the experience of joy) is not present to me. Similarly, the mental states of others, including their thoughts, emotions, and experiences, are primordially inaccessible to us and can only be grasped through what Stein calls the “non-primordial act of empathy” (OPE § 14).

7. The Three Stages of Stein's Theory of Empathy

According to Stein, empathy is a process of being with and feeling with others by directly perceiving their experiences, which Stein describes as the “non-primordial parallel to perception” (OPE §10; also see Dullstein, 2013, p. 349; Svenaeus, 2018, p. 742). Stein refers to empathy as a kind of “fellow feeling” that allows us to understand others by participating in their experiences with them (OPE §14). Stein's theory develops at three stages.

The first stage is the “emergence of the experience” and this involves directly perceiving the other person's embodied experience in an intuitive manner (OPE §10; see also OPE §18; Meneses & Larkin, 2012, p. 157). Unlike the kind of perception found in simulation theories, the Steinian approach involves a particular kind of “seeing” to the extent that one intuits something that does not belong to one's “sphere of ownness” (Shum, 2012, p. 178). When engaging with someone else during this first stage, Stein explains how “I intuitively have before me what they feel. It comes to life in my feeling …” (OPE §18, emphasis added). At this first stage, I am aware that the other person's experiences belong to her and that she is feeling something, which is knowable to me and others within a certain range of indeterminacy.

After initially grasping the other's experience through my own perceptual awareness, the second stage of Stein's theory is a “fulfilling explication” of the object of experience (OPE §10). At this second stage, Stein explains that empathy exhibits the “non-primordial parallel to the having of the experience” (OPE §10). In stage two, I gain a better understanding of the other person's experiences by following through with her in an act of the imagination where I am led by the other. Stage two involves an imaginative “transposal” or “projection” of the self into the other person's experiences. But unlike the projections used by simulation theorists to replicate being the other person to understand her experiences, Stein's projections focus on exploring these experiences as if with the other person as her experience unfolds. Moreover, rather than being conceptualized only as cognitive acts, these projections are experiential, non-intellectual, and intuitive (OPE §21; Meneses & Larkin, 2012, pp. 170, 175–176). During this second stage, there is a shift from an “objectifying intuition” about another person's experiences into a “pre-reflective lived experience” in which one “‘dwells within the Other's experience’” (Shum, 2012, p. 185). Through my non-primordial experience of others, Stein suggests that I feel “led by a primordial one not experienced by me but still there, manifesting itself in my non-primordial experience” (OPE §10, emphasis added). This language that Stein uses of being led by the other, of being “drawn into” (OPE §12), of being “guided by” (OPE §99) and being “pulled … into” (OPE §9) their emotional experiences is made possible through empathy.

After becoming aware of the other person's experiences in stage one, and after being pulled into her phenomenological world where we experience with her and see how her experiences feel to her in stage two, we emerge in stage three with a more comprehensive understanding of the other person's experiences. Stein refers to this third stage as the “comprehensive objectification” of the explained experience (OPE §10; see also Burns, 2017, p. 130; Svenaeus, 2015, p. 241; Svenaeus, 2017, p. 163). During this final stage, the empathizer represents the other person's experience by forming an “intellectual interpretation of what was given of it” (Meneses & Larkin, 2012, p. 175). According to Stein, stage three involves “interpretatively mentalizing” the other person's experience, which involves a higher-level recognition of the other person's primordial experience (Meneses & Larkin, 2012, p. 166; see also Määttä, 2006, pp. 5–6). The result is that it allows individuals to get a better understanding of others’ experiences, which then allows for connections between individuals to occur.

To illustrate Stein's theory of how empathy functions with an example, suppose your friend joyously tells you that she successfully passed her dissertation defence. Stein notes that the joy of the other we experience empathically is numerically the same as the joy we feel firsthand. But the distinction is that it is “a different mode of being given” and thus it is not qualitatively the same as the joy the other person is feeling (OPE §15, emphasis added). For example, if my friend excitedly tells me that she passed her dissertation defence, I can see the joy on her face and her bodily expressions. I can feel happy for her, and I can share her excitement with her. But I do not feel the primordial joy that she is currently experiencing. This is because her feeling of joy is only given to me through the “non-primordial act” of empathy (OPE §14). Although her joy is inaccessible to me, I participate in her experience with her, as this feeling I am having is “primordial as present experience though non-primordial in content” (OPE §9).

Throughout this example of the three-stage process of Stein's theory of empathy, it is important to remember that I am not imposing my own beliefs onto my friend, nor am I attempting to simulate what she is feeling based on how I would react if I were in her situation. Instead, I have a newfound understanding of what this experience means for my friend and I am able to grasp it in a way that recognizes and prioritizes this experience as my friend's experience, not my own. In essence, Stein's account of empathy is helpful for establishing connections between people since we are drawn into their experiences and guided by them as we navigate their experiences with them, not as them. This shift in emphasis from the self to the other is key — particularly in the context of mental healthcare — as it allows for a deeper understanding of the meaning of experience for the other person since it is focused on the other person. Because of this firm insistence on preserving the distinction between the self and the other, Stein's phenomenological theory of empathy provides a reconceptualization of what it means to empathize with others. As a result, not only does her theory of empathy overcome some of the challenges faced by simulation theories of empathy that were discussed above, but there are also several practical advantages of utilizing Stein's theory.

8. The Advantages of Stein's Theory of Empathy in Practice

Although her theory of empathy involves a feeling into the experience of others, and focuses on participating in their experiences with them, one advantage of Stein's theory that distinguishes her view from simulation theories is that her account does not result in a recreation of the other's subjectivity (OPE §12, §16). As Stein argues, the subject of the empathized experience “is not the subject empathizing, but another” (OPE §10). Through acts of empathy, we gain access to another's experiences, but we never take over the person's experience. For instance, if we come across a person who is happy or is grieving a loss of a loved one, we can understand her feelings based on our understanding of the concepts of happiness and grief, respectively. But we can never experience those feelings as experienced by that person from a first-person perspective (OPE §13; Määttä, 2006, p. 5). This is because, as Stein notes, the experiences I feel of the other “does not issue live from my ‘I’” (OPE §10). Rather I have a representation of the other's experience even though I do not experience it myself (Dullstein, 2013, p. 345; Lebech, 2017, p. 113; Svenaeus, 2015, pp. 227, 243).

With respect to members of groups who have been historically marginalized, such as persons diagnosed with schizophrenia or other types of mental illnesses, it is important for persons in positions of power and privilege to not only respect these persons, but they should avoid assimilating, or co-opting, their experiences as much as possible (Molas, 2018, p. 65). On this point, Code argues that empathy at its best preserves yet seeks to know the other person and it respects the boundaries between self and other and does not “seek to assimilate” the other into itself (Code, 1995, p. 141; see also Molas, 2018, p. 65). Although Stein preserves the distinction between the self and other, she does not altogether disregard the similarities. The basic connotation of empathy is that the other person is grasped through an appreciation of similarity (Gallese, 2003, p. 176). However, it remains vital to reiterate that this appreciation of similarity does not require identifying with the other's first-person experiences or overstepping one's epistemic boundaries done in the name of understanding others better.

Furthermore, while it is easier to understand what others are going through if one has had similar experiences, that is not necessary for empathy to occur (Ratcliffe, 2015, pp. 235, 245–246). For example, suppose your best friend is grieving the death of a family member. If you, too, have experienced the grief of losing a loved one, you might be better able to offer the appropriate kind of support to your friend during this difficult time. But acts of empathy do not require us to have had similar experiences in order to understand them. According to Stein's theory of empathy, all that is required is that the other person shares the same fundamental structures of consciousness that makes experiencing possible. And since a person diagnosed with schizophrenia is also an “I” who possesses a living body and perceives the world through her senses, being drawn into her experiences and feeling with her is achievable and is a possibility under Stein's account.

A second advantage of Stein's theory of empathy is that she avoids “projective deception” (Shum, 2012, p. 179; see also OPE §9). Projective deception is the act of ascribing to others mental states that are familiar to us, but that may not reflect their actual experiences. The issue with projective deception is that it privileges the empathizer's own feelings and imposes them onto the other. As Ratcliffe argues, simulation without openness to difference inevitably amounts to “a total failure of empathy; it could not be directed at another person without one's ceasing to experience her as a person at all” (Ratcliffe, 2015, p. 247). Stein explains that by using ourselves as the standard for understanding the experiences and emotional and mental states of others, we “lock ourselves” into the “prison of our individuality” and, as a result, rather than using empathy as a means to gain more insight into the world and others around us, other people “become riddles for us, or still worse, we remodel them into our image …” (OPE §130, emphasis added). The intended purpose or end goal of empathy is not to know all of the experiences of the other. It should instead be conceptualized as a way to engage with others that opens new possibilities of understanding and learning more about them from them and this is precisely what Stein's account permits us to do.

A third advantage of Stein's theory — which separates her conception of empathy from simulation theories — is that she views empathic acts as a joint process. Since all that is required is reconstructing the other person's mental states within our own cognitive systems, simulation should be able to tell us what it is like for another person to experience certain phenomena. Furthermore, because it is a first-person imaginative act, simulation does not necessarily require any interaction with the other person at all. As a result, it can be conceived of as an individual process of attempting to understand others in an indirect manner. By contrast, Stein's theory is an interpersonal process of becoming aware of another's experience in a direct manner. Empathy is always other-focused and thus any attempt to empathize with the other person will be “misguided” to the extent that it takes “first-person replication” of the other person's experience as its goal (Ratcliffe, 2015, p. 231). Thus, as a process of understanding another person's perspective, Stein's account allows caregivers to share in the experiences of those who are diagnosed with mental illnesses to better understand them and this has positive implications for reconceiving therapeutic relationships in beneficial ways.

9. Conclusion

Whereas simulation theories view empathy as a cognitive exercise of the imagination, Stein's approach involves intuitive, affective, and cognitive components for understanding others. Whereas simulation can theoretically be done in isolation and without any direct engagement with others, Stein maintains that empathy is a collaborative and joint-process and is dependent upon the presence of the other for empathy to occur. Finally, whereas simulation theories suggest the possibility of replicating the other's experiences via simulation, Stein's approach allows for direct grasping of the other's primordial experiences but in a non-primordial way.

The role of empathy is to facilitate connections with others and to learn more about them from them. Simulation theories can be useful in some situations, and they can promote understanding in some contexts, but they are not substitutive of the experience of another. Thus, while it is one thing to think about what it is like to be in another person's shoes, thinking about what it might be like is not the same as what it is like for the other person. For as long as empathy is viewed as “putting yourself in someone else's shoes” there is a key experiential element missing that, I argue, the Steinian account can accommodate. At its core, Stein's theory of empathy is preferable to simulation theories within the context of mental healthcare because it focuses on the lived experiences of others. Although it might be the case that caregivers cannot fully understand the primordial experiences of living with schizophrenia, this does not suggest that empathizing with persons diagnosed with schizophrenia is impossible. Rather, it means that current understandings of “empathy” need to be re-worked and Stein's phenomenological theory of empathy offers one potential solution.