By Paul Cruz

Note from JPPE Team: The citations in this paper are in MLA format. This is not the standard Chicago style used by the Journal and future submissions with alternative citation methods will need to be reformatted before publication.

In this paper, I will argue that Edith Stein’s conception of empathy in On the Problem of Empathy can be used to explain the experienced disconnect between neurotypical people and neurodiverse people; between people and other animals; and between people and various robots. In each case, an individual attempts to understand another entity by imagining how they would feel if the other entity’s experience happened to them. The success of this projection depends upon how accurately their imagined reaction aligns with the other entity’s actual reaction, the similarity between the subject and the object, and the feasibility of such a projection.

In section one, I will define Stein’s conception of empathy; in section two, I will apply Stein’s conception of empathy to neurodivergent people and show why there is often an experienced disconnect between neurotypical and neurodivergent people when they try to empathize with each other; in section three, I will explore how certain animals, such as crows, can be empathized with, sometimes to an uncomfortable extent; in section four, I will define the concept of “the uncanny valley” and use Stein’s conception of empathy to attempt to explain why it occurs; in section five, I will address potential counterarguments against my claims; and in section six, I will give my conclusion and closing thoughts.

1: Edith Stein’s Conception of Empathy

It’s important to establish a basic understanding of Edith Stein’s conception of empathy before applying it to the issues explored in this paper. In her work, Stein describes empathy as beginning with an encounter with an external entity. Upon this encounter, the individual intuitively and immediately (“primordially,” in Stein’s words––meaning intrinsically, or arising from within the self) perceives some features and/or actions of the entity as indicative of a broad emotional state.[1] They then project their self emotionally onto the other entity’s perceived or assumed “con-primordial” (Stein 53) (outside of the individual, or extrinsic) emotions, and through projection, they experience their own primordial approximation of that entity’s “con-primordial” emotions. In other words, they experience what they perceive or assume to be the other entity’s emotions through the lens of their own internal emotional framework. As such, the emotions of both entities are experienced primordially within their respective beings. Since those emotions are processed intrinsically through each entity’s respective emotional frameworks, a gap in the resultant emotions is almost––if not entirely––inevitable in natural interactions.

Stein further elaborates on empathy by discussing how the “type” (54-5) of an entity can significantly affect an individual’s ability to empathize with it. Type can be understood as the measure of the degree of similarity a person has to another entity. A greater similarity of type results in a greater amount of potential empathy for the similarly-typed entity, whereas a greater dissimilarity of type can result in a lesser amount of potential empathy for the dissimilarly-typed entity. Type, however, should not be understood as synonymous with mere species, gender, or physical appearance: while all three aforementioned descriptors can influence the degree of type-ness a person has, no single descriptor can determine the degree of type-ness an entity has. Stein uses the example of an injured dog’s paw as something that someone can feel empathy for because the type can be considered similar. While a dog’s paw and a human hand look quite dissimilar, the status of the dog’s paw as being injured allows an individual to empathize with it. This is possible because they can presumably understand what it’s like emotionally to have an injured limb, or, in Stein’s words, they can empathize with “a sensitive limb of a living body” (55), even if said limb is not human.

2: Empathy for the Neurodivergent

With Stein’s conception of empathy and her nuanced idea of type in mind, one can better understand how empathy functions broadly within human interactions. While Stein illustrates cross-species empathy using examples such as the injured dog’s paw, differences in emotional and cognitive type can similarly create empathetic disconnect among humans. Individuals more readily empathize with those who share similar emotional experiences, cognitive processes, or worldviews—essentially, a similarity in cognitive type. Conversely, significant differences in internal emotional or cognitive frameworks naturally form barriers to empathy.

Interactions between neurotypical and neurodivergent individuals, such as those with autism spectrum disorder (ASD) or attention-deficit/hyperactivity disorder (ADHD), illustrate how differences in cognitive type can produce empathetic disconnect. Stein asserts that empathy is fundamentally an imaginative projection of oneself onto another’s "con-primordial" emotional state, presuming similarity in emotional experiences (54-55). Typically, people anticipate emotional alignment unless consistently shown otherwise. Between neurotypical individuals, cognitive-emotional frameworks tend to be sufficiently similar, imparting an empathetic projection that approaches a broad understanding. However, neurodivergent individuals frequently possess distinctive neurological frameworks relative to neurotypical individuals, leading them to interpret and respond to social situations differently. When neurotypical individuals attempt to understand neurodivergent experiences, they inevitably apply their own cognitive frameworks to situations that are processed in fundamentally different ways, creating an intrinsic empathetic gap.

This empathetic gap manifests across multiple domains of cognition, emotion, and social interpretation. Research illuminates important cognitive differences between neurotypical and neurodivergent individuals. Individuals with ASD demonstrate distinctive patterns in several cognitive domains, including theory of mind—the ability to understand that others have different thoughts and feelings than oneself—and information processing (Velikonja et al., 2019). Their "weak central coherence" cognitive style means they often perceive individual details with remarkable clarity while finding it challenging to integrate these details into a cohesive whole. This different way of processing information means that the social situations neurotypical individuals navigate intuitively may appear as disconnected elements to autistic individuals. Similarly, ADHD involves differences in executive functioning, particularly in response inhibition and attention regulation (Martinez et al., 2024).

Emotional experiences also differ significantly between neurotypes. Autistic individuals often experience emotions intensely but struggle with identifying and regulating these emotions, leading to emotional overwhelm (Cibralic et al., 2019). This doesn’t mean autistic individuals feel emotions more strongly in a simple quantitative sense, but rather that they experience emotions in qualitatively different ways than neurotypical individuals. Similarly, individuals with ADHD frequently experience emotional impulsivity—emotions that arise quickly, shift rapidly, and may be expressed with an intensity that neurotypical people find surprising (Soler-Gutiérrez et al., 2023).

Social interpretation adds another dimension to these differences. Autistic individuals often process social information differently, dedicating attention to process aspects of social cognition that neurotypical individuals handle automatically (Masoomi et al., 2024). This means that facial expressions, tone of voice, and other social signals that neurotypical people interpret intuitively may require deliberate analysis for autistic individuals. Individuals with ADHD also experience social interactions differently, with their attention differences and impulsivity creating patterns of social engagement that neurotypical individuals may perceive as unexpected (Morellini et al., 2022; Bora & Pantelis, 2016). For example, a person with ADHD may interrupt conversations, shift topics abruptly, and display intense interest in some topics while seeming disengaged from others. They may also miss subtle social cues and body language if their attention drifts. Such behaviors, while neurologically based, are frequently misinterpreted through a neurotypical lens. Because of empathetic differences in projection, such behaviors are often seen as rudeness, aloofness, or a general personal failing––assumptions that attribute intention to actions more often rooted in fundamental differences of cognition.

Collectively, these differences in cognitive processing, emotional experience, and social interpretation all contribute to a fundamental dissimilarity in Stein’s concept of type. When neurotypical individuals project themselves into neurodivergent experiences following Stein’s model of empathy, they imagine these experiences processed through neurotypical frameworks rather than as they’re actually experienced. This empathy gap is reciprocal: neurodivergent individuals similarly encounter difficulty in projecting themselves onto neurotypical emotional reactions. Autistic and ADHD individuals may perceive neurotypical emotional expressions and social interactions as confusing or unintuitive, amplifying the mutual distance. This reciprocal challenge illustrates how empathy, while originating primordially within each individual, inevitably reflects the degree of cognitive alignment between the empathizer and the empathized. When empathetic projections fall within expected thresholds of understanding, they foster connection; when they fall outside these thresholds, they can create a sense of unease or disconnect that reflects the inherent limitations––and consequences––of projecting oneself onto cognitive frameworks that process the world differently.

3: Empathy for Crows

The concept of type discussed previously also plays a fundamental role in understanding empathy for animals. As illustrated through the injured paw example, individuals can experience empathy for animals by recognizing characteristics that bridge the apparent type-difference between themselves and others––even across species. For any characteristic in an entity that is roughly analogous to something the empathizer can experience––such as with pain in the injured paw example––a person can use that to empathize with such a being.

But discussing only positive examples of empathy risks overlooking a crucial observation: if Stein’s empathy is the projection of oneself onto another’s perceived emotions, individuals should be able to feel a form of negative empathy, too; that is, empathy rooted in emotions that people would consider negative. For example, certain crows in Australia are aware that cane toads are poisonous. Instead of ignoring these toads, the crows are intelligent enough to flip over the frog and stab the frog in the throat, where the flesh is softer and non-poisonous (Bolton). In fact, crows are intelligent to the point that they appear to be both self-aware and capable of retrieving old memories (Begley).

Taken together, these two facts are likely to evoke a sense of unease in the reader. That discomfort, however, is a form of empathy as well: upon considering the previous examples, individuals automatically attempt to project themselves onto the assumed emotions of the mentally-visualized crow killing another being, but the empathy isn’t deeply felt. Rather, they perhaps imagine an instinct-driven creature that lives in the moment to gain sustenance, much like a lion or tiger. The creature is of a fundamentally different type, it would seem, even if they can pull off a few clever tricks to kill things easier. When people realize that crows are deeply intelligent, self-aware beings that are capable of reflecting on the past and planning for the future, the crows' usage of creativity to safely and effectively kill another being, makes the crow become sinister and disturbing. The crow is no longer entirely of a different type than people, but seems, in a way, part human in its consciousness.[2] Crows even remember human faces, and they’ll communicate with each other about the danger a particular human face poses if the crows decide they didn’t like the person that possessed said face (Nijhuis). Even though we, as humans, know that crows are not humans, the fact that their mental faculties seem quite advanced––even similar to our own in some cases––makes us perceive crows as being partially of the same type as us. And in so doing, the previous dismissal of the crow’s behavior no longer becomes entirely possible. We empathize with the perceived emotions of the crows and project ourselves involuntarily yet successfully toward them. We identify with how we think the crows feel, and the cold, matter-of-fact brutality of their actions––the approximation of which is perceived primordially from within us as negative empathy––disturbs us.

4: The Uncanny Valley

This paper has covered the possibilities of empathy among living creatures, whether through positive empathy, negative empathy, type or otherwise. What’s left to be explored now is how empathy responds to something that seems almost alive, but isn’t. First, though, it is necessary to understand the concept of the uncanny valley.

The uncanny valley was first proposed by Masahiro Mori during his tenure as a professor of robotics at the Tokyo Institute of Technology after he wrote an article that was published in an obscure journal in 1970 (Mori). Since then, his ideas have become increasingly popular as technology improves to the point where people can create robots, animations, and figures that approach human likeness without fully achieving it. Mori’s essay postulated that the more human-like a figure becomes in appearance, behavior, and movement, or otherwise, the more affinity one will feel toward it––up to a certain point.

Upon reaching a certain threshold of human-likeness for said figure, the trend of increasing affinity toward it will sharply reverse. The built-up affinity for the figure will quickly become negative. The figure doesn’t evoke a neutral feeling, as humans might have toward a generic industrial machine, but a feeling of visceral disgust, revulsion, and even fear toward the aforementioned figure, greater than that felt towards even a corpse. However, after increasing the human-likeness further and the figure starts to become indistinguishable from a human, the affinity for the figure reaches––and soon after exceeds––the previous high that was achieved just before the prior sharp downturn in affinity. The period in which human affinity toward the aforementioned figure turns negative––and stays negative until the figure looks almost indistinguishable from a human––is called the uncanny valley. Stated differently, the uncanny valley covers the period of immense creepiness wherein a figure––say, a robot––looks almost, but not quite, human.

With the concepts covered in sections one through three of this paper in mind, Stein’s conception of empathy can be used to explain this yet unsolved phenomenon. First, start off by considering a fully industrial robot, devoid of any human-like appearance whatsoever. No matter if all the cogs were polished or broken, observers would not feel empathy for the machine. The robot seems entirely dead to human perception.

As a robot acquires increasingly human-like features, human perception of it shifts: people begin to attribute qualities of consciousness and animation to it proportionate to its human resemblance. However, upon reaching the uncanny valley, the robot loses its accumulated aliveness, without ever becoming quite dead. Observers enter the uncanny valley because they see an object with an appearance that is not quite human yet not quite robotic. The object doesn’t have the aliveness of a human nor the deadness of a pure machine. The object becomes neither alive nor dead, neither human nor robot. It becomes uncategorizable by immediate perception. And because the object fails to be anything distinct, when observers attempt to empathetically project themselves onto such an object—to simulate its emotional state—the projection fails. The object occupies the liminal state between mechanical and human, never achieving the successful emotional mapping that empathy requires.

In failing to project themselves onto said object, humans fail to understand (or at least fail to try to understand) the object. And in failing to try to understand said object, the object becomes utterly alien to the viewer, who begins to see the object as some ambiguous threat. In other words, the failure to empathize with an object makes it uncanny and threatening: it seems completely foreign, yet is unmistakably familiar to the observer; the perceived threat remains there, yet cannot be grasped conceptually.

Implicitly, however, this argument rests upon the assumption that humans can indeed empathize with things that are not alive in any biological sense, as with a robot. And indeed, non-living things can be empathized with: if someone pulls up a picture of an adorable robot, and then thinks about crushing the head of said adorable robot until it violently breaks under the pressure, it seems like an awful thing to contemplate. Despite the robot not being sentient, the appearance of it as something that can be empathized with, particularly in a positive manner, allows the attribution of a sentient-like worth to it. Likewise, if a robot is indistinguishable from a human, people can feel empathy for the robot, because they wouldn’t be able to tell the difference.

5: Potential Objections

Several objections could be raised against the application of Stein’s empathy framework presented in this paper. This section addresses the most substantive challenges to the analysis.

First, critics might argue that the uncanny valley effect stems from reminders of death rather than failed empathy. According to this view, robots that occupy this uncanny threshold between mechanical and human trigger fear of mortality, similar to how corpses might disturb observers.

The problem with this explanation is that it doesn’t align with Mori’s original findings: a corpse produces discomfort, but not nearly as much as entities in the deepest part of the uncanny valley (Mori). If reminders of death were the main cause, corpses should be the most disturbing stimuli––yet in comparison to the furthest depths of the uncanny valley, according to Mori, “We might be glad” to merely encounter “the still valley of the corpse” (Mori).

Second, someone might question how something can be "neither dead nor alive" as suggested about entities in the uncanny valley. After all, some may argue that something is either living or it isn’t.

This objection misunderstands the point being made. While machines are obviously non-living in an ontological sense, what matters for empathy is immediate perception, not logical categorization. When people encounter robots that exist on the border between recognizably mechanical and convincingly human, perception struggles to place them into familiar categories. The emotional response—the basis for empathy—occurs in this immediate perception, not after careful logical analysis. The uncanny valley exploits precisely this gap between perception and categorization.

Third, some may question whether the discomfort experienced when contemplating intelligent crow behavior truly constitutes "negative empathy" as described in this paper. Critics could argue that this reaction is simply fear or disgust at recognizing predatory behavior, rather than a genuine empathetic projection.

This objection fails to account for the key insight that the discomfort emerges specifically when recognizing the crow’s intelligence and self-awareness. The negative feeling doesn’t arise from observing predation generally (as with lions or tigers, who, so far as we know, lack the same degree of cognitive sophistication that crows possess), but specifically when we project ourselves onto a being recognized as possessing qualities of consciousness similar to humans.

The fact that the reaction intensifies upon learning about crow intelligence demonstrates that this is indeed empathetic projection as Stein describes it—the attempt to understand the emotional state of another being that is perceived as partially sharing our type.

6: Conclusion and Final Thoughts

Stein’s conception of empathy demonstrates remarkable versatility, enabling novel interpretations of contemporary phenomena. The framework proves particularly illuminating when applied to the empathetic disconnect between neurotypical and neurodivergent individuals, revealing how differences in cognitive type create fundamental barriers to mutual understanding. Similarly, her theory provides a means to analyze the unsettling nature of empathy toward highly intelligent non-human animals like crows, whose consciousness bears disconcerting similarities to human awareness. Perhaps most notably, Stein’s conception offers a fresh interpretative lens for the uncanny valley—a decades-old problem that continues to challenge researchers.

Despite the significant philosophical depth and avenues for exploration evident in Stein’s work, her ideas remain relatively underexplored compared to those of her contemporaries. Her framework on empathy offers substantive approaches to understanding the complexities of intersubjective experience across diverse domains—from neurodiversity to animal cognition to human-machine interaction. This broad applicability reveals the conceptual richness and contemporary relevance of her philosophical thought.

The application of Stein’s empathy theory to the uncanny valley highlights several areas warranting further investigation. Questions remain regarding the precise mechanisms by which objects transition from being perceived as dead, to alive, to occupying the liminal space of "neither alive nor dead." Additionally, the exact threshold at which an entity becomes perceptually uncategorizable—and subsequently when it becomes re-categorizable after exiting the uncanny valley—deserves deeper examination. These inquiries represent promising directions for future philosophical investigation, demonstrating the enduring applicability of Stein’s conception of empathy to contemporary philosophical questions.

________________

Works Cited

Begley, Sharon. “Brainiacs, Not Birdbrains: Crows Possess Higher Intelligence Long Thought a Primarily Human Attribute.” STAT, 24 Sept. 2020, https://www.statnews.com/2020/09/24/crows-possess-higher-intelligence-long-thought-primarily-human/.

Bolton, Katrina. “Toads Fall Victim to Crows in NT.” Australian Broadcasting Corporation, 14 Sept. 2007, https://www.abc.net.au/news/2007-09-15/toads-fall-victim-to-crows-in-nt/670524.

Bora, E., and C. Pantelis. “Meta-analysis of Social Cognition in Attention-Deficit/Hyperactivity Disorder (ADHD): Comparison with Healthy Controls and Autistic Spectrum Disorder.” Psychological Medicine, vol. 46, no. 4, 2016, pp. 699–716, doi:10.1017/S0033291715002573.

Cibralic, Sara, et al. “A Systematic Review of Emotion Regulation in Children with Autism Spectrum Disorder.” Research in Autism Spectrum Disorders, vol. 68, 2019, p. 101422, doi:10.1016/j.rasd.2019.101422.

Martinez, S., et al. “Unraveling the Spectrum: Overlap, Distinctions, and Nuances of ADHD and ASD in Children.” Frontiers in Psychiatry, vol. 15, 2024, article 1387179, doi:10.3389/fpsyt.2024.1387179.

Masoomi, M., et al. “Emotion Recognition Deficits in Children and Adolescents with Autism Spectrum Disorder: A Comprehensive Meta-analysis of Accuracy and Response Time.” Frontiers in Child & Adolescent Psychiatry, vol. 3, 2025, article 1520854, doi:10.3389/frcha.2024.1520854.

Morellini, L., et al. “Social Cognition in Adult ADHD: A Systematic Review.” Frontiers in Psychology, vol. 13, 2022, article 940445, doi:10.3389/fpsyg.2022.940445.

Mori, Masahiro. “The Uncanny Valley.” Translated by Karl F. MacDorman and Norri Kageki. IEEE Spectrum, 12 June 2012, https://spectrum.ieee.org/the-uncanny-valley.

Nijhuis, Michelle. “Friend or Foe? Crows Never Forget a Face, It Seems.” The New York Times, 25 Aug. 2008, https://www.nytimes.com/2008/08/26/science/26crow.html.

Soler-Gutiérrez, A. M., et al. “Evidence of Emotion Dysregulation as a Core Symptom of Adult ADHD: A Systematic Review.” PLOS ONE, 2023, https://doi.org/10.1371/journal.pone.0280131.

Stein, Edith. On the Problem of Empathy. Translated by Waltraut Stein, 3rd ed., ICS Publications, 1989.

Velikonja, T., et al. “Patterns of Nonsocial and Social Cognitive Functioning in Adults with Autism Spectrum Disorder: A Systematic Review and Meta-analysis.” JAMA Psychiatry, vol. 76, no. 2, 2019, pp. 135–151, doi:10.1001/jamapsychiatry.2018.3645.

________________

[1] By “broad” emotion, I mean an emotion that is understood as belonging to a basic mental category rather than being understood by the specific emotion or emotions that an entity may be feeling. For example, one might intuitively recognize one person frowning and another person sobbing as undergoing emotions belonging to the large emotional category of “sadness,” just as a person smiling might be intuitively recognized as undergoing an emotion belonging to the large emotional category of “happiness”; the emotions are broad enough to fit multiple variations of an emotion under the same general emotional category. That aforementioned person frowning may be experiencing the specific emotion of despair over losing one’s job to a younger, less-qualified worker due to budget cuts because said company is on the verge of going bankrupt while at the same time one still needs to care for one’s dying pet, while the aforementioned person sobbing may be experiencing the specific emotion of stubbing one’s little toe extremely hard while depressed. In both cases, the exact emotion each person is experiencing almost certainly cannot be conveyed by intuitive perception, yet both fall under the broad category of “sadness” in one’s mind, because one can only intuitively understand a portion of another person’s emotions through external perception of features and actions alone.

[2] This does involve a form of anthropomorphization, but it is not a conceptual error within Stein’s framework. Empathy necessarily begins from the subject’s perspective and is filtered through their own primordial cognitive framework. An individual cannot empathically project from a space outside of their own type. As such, the more an individual perceives another being as similar in type, the more that being is interpreted through the subject’s own characteristics.