On Presuppositions

Malcolm Gladwell. Blink: The Power of Thinking Without Thinking. Little, Brown and Company: New York, 2005.

D.L. Rosenhan. “On Being Sane in Insane Places.” Science, vol.179, no.4070, p.250-258 (19 January 1973).

In 1973, the peer-reviewed journal Science published a now-classic study in psychology. In its introduction, the study’s author, D.L. Rosenhan, pointed out that the criteria for determining whether someone is mentally ill are highly subjective. The provocative question Rosenhan asked was this: Do diagnoses of mental illness come from the patient themself, or from the context in which a medical observer examines them? This study was designed to answer that question.

Rosenhan recruited eight normal, healthy individuals, whom he termed “pseudopatients”. Each of the pseudopatients arrived at one of a variety of mental hospitals and checked themselves in, claiming to be hearing voices saying words such as “empty”, “hollow”, and “thud”. Other than this fictitious symptom and the use of pseudonyms, they truthfully answered every question they were asked, and once they were admitted, they immediately stopped complaining of any symptoms at all. At all times, they remained polite and cooperated with the hospital staff (medication was surreptitiously flushed down the toilet). The objective of the study was to see how long it would take for the staff to realize that they were not mentally ill and release them.

The results were shocking. The pseudopatients were all released after between 7 and 52 days, with an average stay of 19 days. But more importantly, not a single one was realized to be sane by any of the doctors or nurses at any of the hospitals. Instead, they were released with diagnoses such as “schizophrenia in remission”. Even more bizarrely, the other patients were more likely to recognize the pseudopatients as normal than the hospital staff. 35 out of 118 genuine patients voiced suspicions about the pseudopatients’ claims of mental illness, and some told them, “You’re not crazy. You’re a journalist, or a professor! You’re checking up on the hospital.”

Even seemingly obvious clues were overlooked. The pseudopatients kept notes on their experience; at first they attempted to conceal these, but soon realized there was no need. Even when the doctors observed them writing, they dismissed it as “note-taking behavior”, another symptom of the presumed psychosis. When they asked ordinary questions such as, “Excuse me, doctor, can you tell me when I am likely to be discharged?”, the staff frequently answered with non sequiturs such as, “Hello, Dave. How are you today?” and then moved off without waiting for a response. In interviews, the patients related their life histories truthfully, which might seem to provide evidence against a diagnosis of mental illness since they were legitimately well. However, these histories were likewise twisted, with psychiatrists interpreting and labeling every drifting apart as “ambivalence”, every fight with loved ones as “outbursts of anger”. There was no evidence of any conscious distortion; the staff legitimately believed in the initial diagnosis, which became a lens that colored all their subsequent interpretation of the evidence. In a sense, the fact that the pseudopatient was in a mental hospital seemed to be all that was required for the doctors and nurses to conclude that they were insane and treat them accordingly. Once a diagnosis was made, it tended to “stick”, and from that moment on, the staff assumed that all behavior stemmed from that diagnosis.

Rosenhan’s paper spurred a variety of reforms in the mental-health industry, and it is to be hoped that the specific problems he identified have been, if not eliminated, at least lessened. But the ugly mark of presupposition has by no means been overcome, even today. This study was about conscious presuppositions, decisions made by doctors and nurses with the full consent of their faculty of reasoning. Chapter 3 of Malcolm Gladwell’s 2005 book Blink discusses a different kind of presupposition – an unconscious kind – that can be brought out by a simple psychological exam known as the Implicit Association Test. Test-takers are shown a list, with two categories, and are asked to place each item onto the list in its appropriate category as quickly as possible. For example, consider this list:

Male Female
John
Bob
Amy
Holly
Joan
Derek

Most participants find this test easy. Now consider a minor modification, where each category consists of two possibilities:

Male or Career Female or Family
Lisa
Matt
Laundry
Entrepreneur
John
Merchant
Housework

Though this test is more difficult, most test-takers still do fairly well on it, consistently answering in between four and six hundred milliseconds. But now, consider a subtle variation:

Male or Family Female or Career
Lisa
Matt
Laundry
Entrepreneur
John
Merchant
Housework

When presented with this altered version, most participants – of both genders – score significantly worse, both taking longer to assign items (up to three hundred milliseconds longer, a huge margin in a test of this type). Evidently, whether on a conscious level or not, most people associate careers and business with men and housework and family with women. Though we can overcome these associations by an effort of will, it is considerably more difficult and requires more cognitive effort. Despite decades of feminist advocacy for women’s equality, despite laws and policies that combat gender discrimination, it seems that these prejudices have not been fully eradicated, even among the best of us.

Now consider the most unsettling implicit association test of all:

European-American
or Good
  African-American
or Bad
Hurt
Evil
Glorious
Wonderful

European-American
or Bad
  African-American
or Good
Hurt
Evil
Glorious
Wonderful

The results of this test are frightening: No matter what participants say about their attitude toward racial equality, a large majority – about 80% in general, and about half of African-American participants – take measurably longer when asked to link positive concepts with African-Americans than with Caucasians. Again, it seems that despite everything that has been said or written on the subject of racial equality, at some level the old biases persist. Even people who do not consciously harbor any racial prejudice, who would never in their lives dream of discriminating against another based on gender or skin color, are often not able to fully overcome it. From somewhere – and it is an open question from where these biases were learned – the vile memes are still being transmitted. (For those of you who want to test your own preconceptions, the Implicit Association Test can be taken online, at https://implicit.harvard.edu/implicit/demo/measureyourattitudes.html.)

Granted, whether any of these implicit associations translate into any measurable difference in our actual behavior has not been determined. But these results should still be deeply disturbing to any educated, tolerant individual: ugly, harmful presuppositions can take root despite our best efforts to eradicate them.

Together, these experiments show just how vulnerable the human mind is to this type of error. Information learned both consciously and unconsciously can seep in to bias our decisions without our being aware of it, and in spite of our attempts to correct for it. Even highly educated, tolerant people, the kind of people that would probably be considered least likely to hold presuppositions, can be vulnerable.

How then can we believe any of our positions to be reliable? Is the quest for knowledge hopelessly subjective? The next post in the Observatory will consider these important questions and suggest a solution.

About Adam Lee

Adam Lee is an atheist writer and speaker living in New York City. His new novel, Broken Ring, is available in paperback and e-book. Read his full bio, or follow him on Twitter.

  • tobe38

    Excellent post, both thoughtful and thought provoking.

    In response to your last point about subconscious racial biases in even the most well-meaning people, could the stigmas of ‘political correctness’ be a factor in these disturbing results?

    I believe that many people are truly not racist, neither consciously nor unconsciously, but in the ‘PC’ age almost everyone is afraid of being perceived as racist, because even casual use of the word “black” (as in, “I’d like a black coffee) can provoke a recoil of disgust from the PC brigade. In short, people who are not racist may hesitate to answer not because they are unconsciously prejudiced, but because unconsciously they want to take care that their answers won’t be perceived as racist.

    I could rant for hours about political correctness, but I won’t here because I hope I’ve made my point in sufficient detail. All I will say is that I rigorously oppose Political Correctness. Instead of pretending that we are all little clones of each other, we should be recognizing, acknowledging and maybe even, Satan forbid, celebrating the differences between us. The Christians say don’t judge or you’ll be judged yourself. I say judge to your heart’s content, but fairly and by the right criteria.

  • BlackWizardMagus

    I just wonder if these differences, while they seem to be a huge margin in this type of test, really create any sort of statistically noticeable trends. I mean, it takes us just a bit longer to connect women with work (which isn’t even bias; it’s also just plain history), so what? We could theorize of the effects this might have, but that’s not what this test shows, so it’s really just a point of interest for now.

  • Alex

    I assume someone thought to swap all the words around (IE, female on the left, male on the right in the first one)? I wonder if part of the problem might be the learned habit of reading left to right and a tendency to think of these words in a certain order (“male and female” being less phonetically awkward than “female and male”).

  • http://www.patheos.com/blogs/daylightatheism/ Ebonmuse

    In short, people who are not racist may hesitate to answer not because they are unconsciously prejudiced, but because unconsciously they want to take care that their answers won’t be perceived as racist.

    I wonder if part of the problem might be the learned habit of reading left to right and a tendency to think of these words in a certain order (“male and female” being less phonetically awkward than “female and male”).

    These are both interesting suggestions to explain these results – I don’t actually know whether the word order is reversed from test to test, though it certainly should be, to exclude that as a possible source of error. Even if not, though, one might argue that viewing “male and female” as more natural than “female and male” may itself be a type of that subtle subconscious bias this study was designed to measure.

    As BlackWizardMagus pointed out, it’s an open question whether this phenomenon actually affects how people treat each other consciously. But there’s no question that subtle factors can affect our decisions in ways we’re not explicitly aware of. Gladwell points out in the same chapter of Blink that 58% of CEOs are six feet tall or more, while only 15% of men in general are that tall. Even more striking, only 4% of men are six foot two or taller, but among CEOs, it’s almost one-third. Clearly there is a subconscious bias here (an irrational belief that tall people make better leaders) that is affecting conscious decision-making in a strikingly obvious way. We don’t yet know if implicit racial or gender bias affects decisions in a similar fashion, but it would be foolish to dismiss the possibility.

  • http://www.reluctantatheist.com Reluctant Atheist

    I rather like Alex’s thought about reading left to right.
    The 1 part I balk at, is when I looked at the ‘European-American = Good, African-American = bad,” I took a few milliseconds (or whatever amount of time) to think to myself, “Neither are bad.”
    It’s this either/or thing. People think in terms of absolutes: it’s either all good, or all bad. Life ain’t like that.
    I took 1 of those ‘Racial’ tests, where the length of time is measured, you click on the left for identifying Asians, on the right for Whites, & they change it around several times (Right for asians, left for whites, reversed, reversed again, etc).
    All it proved is that I have trouble switching columns. That’s it. I made a LOT of mistakes, on both sides, to the point where my results were inconclusive.
    You can’t encapsulate how a person is based on some 5 minute test that moves the goalposts.
    I tend to stop & think before I answer questions.
    Maybe that’s just me.

  • Alexander P

    I’ve seen those association tests and I’m skeptical of their usefullness. Is it a reasonable conclusion that people are still harboring biases in a sub-conscious level? I see one possible source of error. When I take the first test, I automatically establish the pattern of yes and no in my mind such as “yes yes no no no” which endure in my memory. Then with the next test which has the same word order but different associatied terms. I’m still remembering the my response pattern of the previous test so it takes me a few moments to make the switch from the previous association to the next association. Once I’ve made that mental switch, I can proceed normally.

    I don’t think your conclusions are valid at all, it’s just establishing a memorized rote which a person follows. I’d be more impressed if the word order were actually randomized and isn’t influenced by the memory of the previous test.

  • BlackWizardMagus

    As well, I took the Native American one, and I got screwed up because it was white on the left, NA on the right. Then it was domestic names on the right, and foriegn on the left. Then the added them together on the same sides, but then switched the W and NA, so that after constantly putting whites an white city names on the left, I was suddenly having to reverse. Quite frankly, the test led me to associate whites with white cities with the left side, then switched it and said it was my fault. I think that these tests are looking for something so small and specific that it’s impossible to find; no test is ever perfect, and the margin of error is too small to allow for mistakes and still be conclusive. We definately do have unconcious biases, but I think this test is a bad measure.

  • http://www.patheos.com/blogs/daylightatheism/ Ebonmuse

    I checked on the IAT website, and the order of presentation does seem to be randomized. Sometimes a category will be on the left, sometimes on the right, and the order in which items are presented also varies each time a new test is started.

    The point as I understand it is, if people were just repeating previously learned patterns, then why would the results in general skew so dramatically toward a particular set of associations? Since the category order is randomized, if that was the error people were making, the associations should be randomly distributed. But they’re not. There is a consistent association, across many test-takers, between certain items and categories.

  • BlackWizardMagus

    But the question is inherently going to have that issue; they specifically have you do the “common” associations, then do the opposite (I thought, anyway). Well, this is naturally going to first bias you to the common association, and screw up your reactions because you, in part at least, asociating “white” with “left”, not with “America”. I guess it’s like a middle step that might be the real player. I wouldn’t be surprised if there was still some bias, but I think that it’s going to be less clear overall for technical reasons.

  • ebohlman

    In his classic The Nature of Prejudice, Gordon Allport described certain nouns and adjectives as “labels of primary potency”; basically, such a label is capable of hijacking most people’s cognitive processes, causing them to view just about every aspect of the person in terms of the label. This is a case of the phenomenon called “group attribution error” (GAE); we tend to explain the behavior of a member of a group that we don’t belong to, particularly if it’s one we’re “distant” from, in terms of relatively fixed characteristics of the group itself, whereas we’re more likely to explain the behavior of members of our own group in terms of individual or circumstantial factors. A related cognitive bias is the “assumption of outgroup homogeneity” in which members of a “distant” group are treated as some sort of Borg-like collective.

  • Archi Medez

    Re: False positive diagnoses of schizophrenia. This partly reflects the lack of good tests for schizophrenia, including lack of neuroscientific tests now used. Also, even back when that study was done, it appears that the doctors and nurses did not spend much time observing the pseudopatients. Nevertheless, this was certainly an eye-opening study.

    Re: Interpretation of reaction time results. Reaction time is one of the most ambiguous measures (and among the most frequently used) in psychology (that’s not just my opinion, but also that of R. D. Luce in regard ot reaction time results generally). It is not clear, for example, whether what was shown was indicative of prejudice, or knowledge of prejudice, self-awareness of being tested for prejudice (which is very obvious–in fact, far too obvious…this artifact alone brings the interpretation of the results into doubt). It is necessary to obtain other measures in conjunction with reaction time in order to get greater confidence that what is being measured truly is what is intended to be measured. (This is in addition to having a large number of different conditions, which help to rule out alternative explanations…but alas this becomes increasingly and often prohibitively expensive). In addition, other tasks should be used to show that the effect is not specific to the type of task in question. (Other ways to ensure generalizability of the results: test different age groups, test different populations (i.e., test people in different countries).

    Re: Height of CEO’s. The irrational belief that taller people make better leaders may indeed be a factor here. But there are a bunch or intercorrelated factors here, eg., height is correlated with socio-economic status (SES); SES is correlated with education; one’s SES is associated with one’s parents’ SES; height is correlated with parents’ height; SES is correlated with nutrition; nutrition is correlated with height; education is correlated with nutrition; education is correlated with IQ…and there’s probably more I’m leaving out. I’m not sure the research has been done to sort out the roles of all of these factors determining CEO’s heights. Confidence and dominant style of social interaction might be related to height insofar as playground, high school, and college socialization may establish this pattern. Yet another factor to be considered is women’s preference for taller males, which has been shown in 37 out of 37 cultures studied. (Many opportunities for preferential treatment could be affected by this bias).

    Some of our biases are truly staggering; by my above cautions regarding methodology I don’t mean to downplay the biases or deny that they could exist, of course. For example, people will give a higher grade to an essay attributed to an author judged to be good-looking vs one judged to be not good-looking. (The same essay is assigned to different groups, with a picture of the author on the front. E.g., an essay would be assigned with a good-looking author’s picture to one group, and the same essay would be assigned with a not-good-looking author’s picture. The groups then were given a set of criteria and were asked to grade the essays accordingly. They were not asked to consider anything else such as the author’s appearance; appearance wasn’t mentioned. Other judges had previously judged the pictures independently of the essays to obtain the “attractiveness” ratings). If memory serves me correctly, there was actually a 10% difference in the average marks, with the advantage going to the good-looking authors.

  • Archi Medez

    Note: In the above example I just cited, the people grading the essays did not know about any attractiveness ratings, and were not aware that author’s attractiveness was being studied.

  • Archi Medez

    Another note: I actually meant to say economic status (ES) where I said socio-economic status (SES) above. SES is usually a combined variable that often includes factors such as education; the definition can vary depending on what variables the researcher includes.