A Critique of the Empathy Quotient (EQ) Test: Introduction and Part 1

The Empathy Quotient (EQ) test was designed by Simon Baron-Cohen and Sally Wheelwright, and is included in their 2004 paper The Empathy Quotient: An Investigation of Adults with Asperger’s Syndrome or High-Functioning Autism, and Normal Sex Differences. It is frequently used as one of the primary measures of empathy in autistic people, and is often brought forward to support the twin contentions that a) autistic people have below-average levels of empathy and that b) autism is, by definition, a low-empathy condition.

The test consists of a series of 60 statements, to which the respondent must answer one of the following:

  • Definitely agree
  • Slightly agree
  • Slightly disagree
  • Definitely disagree

The resulting test scores are interpreted in the following ways:

  • 0 – 32 = low (most people with Asperger Syndrome or high-functioning autism score about 20)
  • 33 – 52 = average (most women score about 47 and most men score about 42)
  • 53 – 63 is above average
  • 64 – 80 is very high
  • 80 is maximum

As is standard for questionnaires and psychological tests, there are a number of “filler statements” that have nothing to do with the purpose of the test. On the EQ test, twenty filler statements are inserted, in the words of the authors, “to distract the participant from a relentless focus on empathy.” (Baron-Cohen and Wheelwright, 166)

I’ve gone through all 60 statements on the EQ test and attempted to categorize them under the following headings:

  • Filler statements
  • Statements regarding cognitive empathy
  • Statements regarding emotional/affective empathy

As far as I can tell, statements 2, 3, 5, 7, 9, 13, 16, 17, 20, 23, 24, 30, 31, 33, 40, 45, 47, 51, 53, and 56 are the filler statements. I won’t be critiquing them, since they have no impact on the EQ score.

Regarding the other two categories, the authors are quick to point out that, when setting up the test, they attempted to make a distinction between statements designed to measure cognitive empathy and statements designed to measure emotional/affective empathy, but gave up on the effort because there is so much overlap. I am very cognizant of the complex nature of the overlap, but I’ve separated these statements out, mainly for the purpose organizing my critique. In the category of cognitive empathy, I have also separated the statements about reading nonverbal cues from the statements about perspective taking.

The critique consists of these components:

The Introduction provide a discussion of the basics of the EQ test.

Part 1 provides a definition of cognitive empathy, along with a critique of the statements on the EQ test concerning cognitive empathy and nonverbal cues.

Part 2 examines the statements on the EQ test that cover cognitive empathy and perspective taking.

Part 3 provides a definition of emotional/affective empathy and includes a consideration of the statements on the EQ test that speak to this form of empathy

The Conclusion brings together my thoughts about the general nature of the test and its implications for autistic people.

Part 1
Of the 40 statements geared toward measuring empathy on the EQ test, the vast majority – 27 – have primarily to do with cognitive empathy. Of course, some of these statements encompass both cognitive and emotional components, but in them, a lack of cognitive empathy is an implicit explanation for the lack of a normative emotional response, so I have included them under the cognitive empathy heading.

In The Empathy Quotient, Baron-Cohen and Wheelwright draw on a definition of cognitive empathy as “using a ‘theory of mind’ (Astington, Harris, & Olson, 1988; Wellman, 1990) or ‘mindreading’ (Baron-Cohen, 1995; Whiten, 1991).” According to the authors, cognitive empathy encompasses “setting aside one’s own current perspective, attributing a mental state (or ‘attitude’) to the other person (Leslie, 1987), and then inferring the likely content of their mental state, given the experience of that person.” (Baron-Cohen and Wheelwright, 164)

For those not familiar with the term “theory of mind (ToM),” Baron-Cohen defines it in the following way in his 2001 paper Theory of mind in normal development and autism:

A theory of mind remains one of the quintessential abilities that makes us human (Whiten, 1993). By theory of mind we mean being able to infer the full range of mental states (beliefs, desires, intentions, imagination, emotions, etc.) that cause action. In brief, having a theory of mind is to be able to reflect on the contents of one’s own and other’s minds. Difficulty in understanding other minds is a core cognitive feature of autism spectrum conditions. The theory of mind difficulties seem to be universal among such individuals.” (Baron-Cohen, 3)

I’m including the preceding paragraph not only for purposes of definition, but also to illustrate a) Baron-Cohen’s assumption that autistic people lack a ToM, and b) to make clear the rather dire consequences of this conclusion for autistic people — that is, that we lack one of the essential qualities of full humanity. Because the definition of cognitive empathy in use on the EQ test is based on an equivalence with ToM, and because Baron-Cohen considers ToM a quintessential component of humanity, it’s vitally important to critique the sections of the EQ test that contribute to his conclusions about cognitive empathy and autism.

I want to point out that the definition of cognitive empathy being used in Baron-Cohen and Wheelwright’s paper is quite different from the one that I have been using for some time. In my understanding, cognitive empathy has to do with being able to read nonverbal cues (body language, facial expressions, the expressions in the eyes, and so on) in order to intuitively “tune in” to what another person is thinking or feeling. I have not been using it simply to cover being able to see things from another person’s perspective or to understand the other person’s mental state.

To me, these are two separate, albeit related, processes. I have difficulty reading the nonverbal cues of non-autistic people, but I can’t remember a time in my life that I didn’t ask numerous questions or make numerous observations in order to understand the perspectives of other people; and I certainly can’t recall ever making the assumption other people’s thoughts and feelings were exactly like my own in every instance. In fact, my perception that my family members had values, and perspectives, and thoughts, and feelings that were altogether different from my own engendered a deep sense of aloneness in me from the time I was very young. Feeling like a stranger in a strange land is common for autistic people; the sense of being an anthropologist from Mars is a reflection of the fact that we are often keenly aware that other people perceive the world in ways vastly different from our own, and that we seek to make sense of it.

Given that I consider the reading of nonverbal cues and the ability to understand the perspective of others two separate processes, I will speak to the statements concerning them separately.

Statements that measure being able to read nonverbal cues
Here are the 15 statements on the EQ test that measure the respondent’s ability to pick up nonverbal cues:

1. I can easily tell if someone else wants to enter a conversation.
8. I find it hard to know what to do in a social situation.
10. People often tell me that I went too far in driving my point home in a discussion.
14. I often find it difficult to judge if something is rude or polite.
19. I can pick up quickly if someone says one thing but means another.
26. I am quick to spot when someone in a group is feeling awkward or uncomfortable.
35. I don’t tend to find social situations confusing.
41. I can easily tell if someone else is interested or bored with what I am saying.
44. I can sense if I am intruding, even if the other person doesn’t tell me.
46. People sometimes tell me that I have gone too far with teasing.
52. I can tune in to how someone else feels rapidly and intuitively.
54. I can easily work out what another person might want to talk about.
55. I can tell if someone is masking their true emotion.
57. I don’t consciously work out the rules of social situations.
58. I am good at predicting what someone will do.

I’ll begin by pointing out the inherent biases of these statements. They were clearly written by non-autistic people, with the assumption that the person being observed by the respondent is non-autistic, and that the social settings to which they refer are composed mainly of non-autistic people. (For example, the statement “I find it hard to know what to do in a social situation,” assumes a conventional social situation in which most, if not all, of the other people are non-autistic.) In other words, the statements are created by “normal” people, to measure responses to “normal” people, in “normal” settings.

When it comes to measuring empathy, this bias is a significantly troubling one — not just for autistic people, but for disabled people in general. The statements do not come from the perspective of autistic/disabled experience, they do not measure the respondent’s ability to read the nonverbal expression of autistic/disabled people, they do not consider the social position of autistic/disabled people in conventional social settings, and they do not consider any settings in which autistic/disabled people are the majority members.

To understand the implications of this bias, consider the first statement: “I can easily tell if someone else wants to enter a conversation.” As an autistic person, when I am in a “normal” social situation, I have great difficulty knowing when to jump into a conversation, and I am mystified by the fact that others seem to be reading one another’s signals and knowing when to let one another in. (In settings with autistic people, I do not have similar difficulties, as I understand both the cues and the social norms much better.) So, I would likely answer “Strongly disagree” to the first statement, simply because most situations in which I find myself involve “normal” people, who put out cues I do not understand; my answer, based solely on my minority status, would contribute to a lower empathy score. (I could skew the results by imagining myself only in situations with autistic people, but since the test is clearly measuring what happens in normative situations, I would respond to the statement based on the totality of my experience.)

Because the people writing the test are non-autistic, they have no idea of the methods that I use to work around the problem of being unable to read “normal” social cues. In instances in which I cannot intuitively tell when someone wants to enter a conversation, I tend to consciously look for people who aren’t able to get a word in edgewise, and I attempt to make room for them. In terms of perspective taking, this approach shows a significant level of cognitive empathy: I observe process, I see who is being excluded, and I identify with the experience of exclusion to such a degree that I attempt to ease the discomfort of other people. The fact that the authors of the test do not understand my adaptive mechanisms is quite problematic, because while my inability to tell when “normal” people want to enter a conversation would contribute to a low score, my adaptive mechanisms reflect a high level of cognitive empathy that the test does not pick up.

The statement about knowing when to include others in a conversation also fails to address the issue of what happens to autistic or otherwise disabled people in “normal” social settings. Given the social roles in which disabled people tend to be cast, this omission is a serious one. Disabled people often find ourselves wanting to enter a conversation in a social setting, only to have other people exclude us completely. I have been in a number of situations in which I’ve had this experience. “Normal” people were unable to read my nonverbal signals sufficiently to bring me in; in fact, they rendered me socially invisible. I always hesitate to talk in universals, but this experience is about as close to a universal one as you can find for disabled people, and anyone familiar with both the experience and the sociology of disability easily understands it.

I’m certain that if you asked most “normal” people whether they chronically fail to notice when disabled people want to enter a conversation, they’d deny it. For the most part, they pay so little attention to us that they probably don’t even realize what they’re doing. But these are the very same people who would very likely answer “Strongly agree” in response to the statement that they can easily tell if a person wants to enter a conversation. And the only reason that, according to the test, such a response is valid is because, in most instances, such people actually do notice other people sufficiently to read their signals. Thus, all the response indicates is that people in the majority are attentive to other people in the majority. It does not address a bias against disabled people, in the same setting, that is based on anything but empathy.

Finally, all of the statements that cover one’s ability to decipher the nonverbal cues of “normal” people rest on the assumption that everyone should be able to intuitively do so, and that an inability to do so is evidence of a lack of empathy. For example, the statement “I am quick to spot when someone in a group is feeling awkward or uncomfortable,” assumes that the respondent is looking at a non-autistic person. In this instance, I can certainly see how it might be difficult for an autistic person to quickly spot whether a non-autistic feels awkward or uncomfortable, because of difficulties in reading the person’s cues. I can also see how it might be easy for a non-autistic person to quickly spot whether another non-autistic person feels awkward or uncomfortable, because of an understanding of those same cues.

But of course, the test does not assume that the person being observed is autistic, that everyone should intuitively be able to read the nonverbal cues of the autistic person, and that an inability to do so is evidence of a lack of empathy. After all, if the statement about intuitively reading awkwardness or discomfort assumed that the respondent were looking at an autistic person, the results would come out quite differently, for two reasons: a) autistic people stand a better chance of reading one another’s signals properly, and b) non-autistic people usually find it very difficult to read autistic people’s signals properly.

For example, when I am in a store in which very loud music is playing, I have never had the experience of a non-autistic person being able to read my discomfort or note my awkwardness. Not once. Not ever. And yet, for me (and for a great many other autistic people), being in a store with very loud music is the hell-realm, and the question of whether to stay or go, whether to ask the store manager to turn down the music or not, whether to cry with frustration or put my fingers in my ears, places me in an extremely awkward position. My experience surpasses “normal” social awkwardness and “normal” social discomfort by several orders of magnitude, and yet non-autistic people fail to intuitively recognize that I’m having any kind of aversive experience at all. In every such situation I enter, I have to explain my experience, in detail, if I am to stand a chance of someone responding appropriately.

In general, when it comes to their relationships with autistic people, most non-autistics cannot, in the language of statement 52, “tune in to how someone else feels rapidly and intuitively.” And yet, of course, no one considers neurotypicality to be, by definition, a low-empathy condition.

Next: In Part 2, I will turn to the issue of perspective taking.

Rachel Cohen-Rottenberg blogs at Journeys with Autism, and presides at Autism and Empathy.

A Critique of the Empathy Quotient (EQ) Test: Introduction and Part 1 appears here by permission.

Rachel Cohen-Rottenberg’s Memoir is The Uncharted Path.

[image via Flickr/Creative Commons]

on 08/30/11 in Autism, featured | No Comments | Read More

Leave a Reply