The Complexity Critique

Razib at Gene Expression:

I have always been struck by starkness of human hypocrisy and its incongruity in the face of avowed beliefs. … Sin is common, and human weakness in the face of contradiction the norm.  Mens’ hearts are easily divided, and simultaneously sincere in their inclinations. … All this leads to the point that I believe far too many of those of us who wish to comprehend human nature scientifically lack a basic grasp of it intuitively. … Many atheists simply lack a deep understanding of what drives people to be religious, and that our psychological model of those who believe in gods is extremely suspect. The "irrationality" and "contradiction" of human behavior may be rendered far more systematically coherent simply by adding more parameters into the model. … When I engage with these sorts of issues with readers of Overcoming Bias or Singularitarians my suspicions become even stronger because I see in some individuals an even greater lack of fluency in normal cognition than my own. … My point is that understanding human nature is not a matter of fitting humanity to our expectations and wishes, but modeling it as it is, whether one thinks that that nature is irrational or not within one’s normative framework.

This frustrating critique is frustrating common: "You’re wrong because your model is too simple.  But I’m not going to tell you what your model is missing, at least not in a clear enough way to help you improve your model."  Yes of course almost all our models are too simple.  We all know that; what we don’t know is exactly what complexities we should be adding to our models.  And for the record I was a teen cultist and my dad and brother were/are church pastors.

For social scientists I think there is actually an advantage in having a less powerful intuitive understanding of human behavior – it helps us notice things that need explaining.  To want to explain particular human behaviors you first need to see them as puzzling, and people with powerful intuitive understandings can predict behavior so well intuitively that they often don’t notice behaviors that are at odds with our best theories. 

GD Star Rating
loading...
Tagged as:
Trackback URL:
  • http://anittahpatrick.com Anittah Patrick

    What?!?! Erroneous.

    Having a deeper understanding of human behavior, whether born from intuition or thousands of dollars in cognitive behavioral therapy, does not necessarily mean you’re at a disadvantage when it comes to the critiques of models of human behavior.

    In fact, when reading a reductive explanation for why we humans do what we do, being able to see the delta between _how the model says people act_ v. _how people actually act_ would be something you’d need to have prior to articulating a critique. If lacking the knowledge of _how people actually act_, then there is no baseline against which the shoddy model can be vetted against.

    This is why we need more ladiessss studying judgment and decision making, IMHO. Because as Ariely intimated in the chapter involving self-gratifying undergrads, men are fairly simple creatures.

  • billswift

    Models are always simpler than what is modelled. If it isn’t simpler it is a copy, not a model. It’s npot a frustrating critique, it’s a worthless one; It tells you nothing you didn’t already know.

    Understanding others is useful to my program of improving myself, but removing barriers to clear thought in my own thinking is more directly useful, right now.

  • Tom

    In fact, when reading a reductive explanation for why we humans do what we do, being able to see the delta between _how the model says people act_ v. _how people actually act_ would be something you’d need to have prior to articulating a critique. If lacking the knowledge of _how people actually act_, then there is no baseline against which the shoddy model can be vetted against.

    Around here we use data.

  • http://yudkowsky.net/ Eliezer Yudkowsky

    I would begin by asking the question, “What is an experimental prediction that you think you make differently from me?”

  • http://profile.typekey.com/aroneus/ Aron

    Some good quotes here for the Asperger’s website on career suggestions.

  • http://occludedsun.wordpress.com Caledonian

    My experience is that people who complain about the ‘simplicity’ of the explanations for various human behaviors are often reacting to the implicit suggestion that people who engage in those behaviors are stupid, foolish, or motivated by unreasonable emotions.

    In some cases, it’s simply that making an attempt to be rational and objective about the behaviors causes people to abandon them, and those who have an interest in propagating or encouraging the behaviors take a dim view of that outcome for obvious reasons.

  • http://www.hopeanon.typepad.com Hopefully Anonymous

    I think Razib has a fundamentally good point about the distinction between scientific inquiry and the attempt to reduce irrational behavior in society. One can study religion or bias in society without making normative judgments. To that extent at least, Razib’s post is a helpful criticism of some readers of Overcoming Bias and some singulitarians, who don’t seem to get much deeper in their analysis of apparently more biased people than saying “less rational than me, bad!”

  • http://www.scienceblogs.com/gnxp razib

    For social scientists I think there is actually an advantage in having a less powerful intuitive understanding of human behavior – it helps us notice things that need explaining.

    yes. this may be so.

    My experience is that people who complain about the ‘simplicity’ of the explanations for various human behaviors are often reacting to the implicit suggestion that people who engage in those behaviors are stupid, foolish, or motivated by unreasonable emotions.

    right, but that’s not me.

    “What is an experimental prediction that you think you make differently from me?”

    to be explicit about it may atheists have a view of religion which mirrors their own rational models of the world premised on clear and distinct axioms. some religionists, generally the professionals and “religon nerds,” are also invested in this model. many religionists will agree with the characterization that their religion is an operation of inferences from a particular set of premises (e.g., the lord’s prayer, the bible, the vedas, etc.). i think these assumptions lead to false predictions.

    and yes, social science is the light in the darkness of our ignorance.

  • http://www.scienceblogs.com/gnxp razib

    and to be explicit, i do much prefer the company of OB readers and singulitarians than i do “normals.” the set of axioms which i have intersects with this group to a great extent and the clarity of the reasoning process is satisfying. that being said, i have seen in my own confusion as to the behavior of sincere religious people an echo of the tendency to project onto to other humans my own bias toward systematic coherency.

  • http://www.scienceblogs.com/gnxp razib

    also, how i view religion, is a very abbreviated post which outlines the general outlines of my mental model of religion.

    And for the record I was a teen cultist and my dad and brother were/are church pastors.

    one hypothesis which i think needs exploration is the possibility that bracketed into the term “religious” are people with very different cognitive outlooks and frames who trade in the same nominal semantic currency and so are not aware of the very radical difference in their cognitive gestalt when it comes to the religion which they presumably share. i have a hunch that many highly textual and theologically oriented religionists are actually much more like atheists than they are like the vast majority of religious people. i also suspect that a disproportionate number of religious professionals and writers of books outline the nature of religion in an apologetic manner fall into this category.

  • Dagon

    Don’t say “You’re wrong because your model is too simple”. How do you know what’s too simple? Say “you’re wrong because you mispredict X”.

    It seems unlikely that very many models are too simple. It seems like they’re probably too complicated, in fact. In many cases, they’re trying to explain things on the wrong level and have incorrect unstated assumptions.

    Much like the earth-centric models of astronomy aren’t too simple – they’re way more complicated than the heliocentric gravitational model.

  • no one

    It’s also true that “bracketed into the term ‘atheist’ are people with very different cognitive outlooks and frames …”. It is true pretty much any time you group people together based on qualities other than “having the same cognitive outlook and frames …”. Isn’t this obvious, common knowledge?

    Only people who know nothing about religion think that religion is nothing but belief (i.e., doesn’t include shared rituals, morality, history, worldview, prejudices & biases, culture, etc.).

  • http://www.scienceblogs.com/gnxp razib

    Only people who know nothing about religion think that religion is nothing but belief

    see my interlocutor “Reality Bites” on the linked post. this is not atypical. many non-religious people have a tendency of reducing religion to one dimension or parameter. e.g., “religion is about group identity.” “religious is about belief.” “religious is about fear of death.”

  • no one

    I agree that this sort of foolishness happens a lot, but I think it’s more a case of many people have a tendency to reduce multi-dimensional concepts to one dimension. It has no particular relation to religion. It’s one of many general cognitive tendencies, just like the tendency to prefer anecdotal data from acquaintances over much larger amounts of aggregate (albeit faceless, impersonal) data.

    Of course, as others have noted, reduction is absolutely necessary in most interesting cases, and is highly desirable when done properly. It’s very rare though that reducing to just one dimension yields a useful model that makes good general predictions. It might yield good predictions and have good explanatory power in a narrow domain though, so it’s not necessarily always a bad thing, but to assume that it is a good general model because it captures one important dimension is sheer idiocy.

    On a related note, I wish that Harris’ and Dawkins’ had explicitly said (in their respective works), “this is my simplified model of religion which I recognize is very different from the model that most religious people have but which I believe (for reasons discussed below …) is a good model for the aspects of religion and its relation to society that I am going to treat.” I think they could make a good argument to that effect, which would obviate the many objections that they are ignorant of religion (which they are) and get those people who ignore them because they think their model is oversimplified to actually engage with the valid points they are making, many of which do just concern belief and the consequences of belief.

  • http://www.scienceblogs.com/gnxp razib

    On a related note, I wish that Harris’ and Dawkins’ had explicitly said (in their respective works), “this is my simplified model of religion which I recognize is very different from the model that most religious people have but which I believe (for reasons discussed below …) is a good model for the aspects of religion and its relation to society that I am going to treat.”

    dawkins comes close to this in the god delusion. in fact, the first half the book is a relatively competent examination of various dimensions. the second half just pretends though that it is one dimension for the purposes of refutation.

  • http://www.scienceblogs.com/gnxp razib

    I agree that this sort of foolishness happens a lot, but I think it’s more a case of many people have a tendency to reduce multi-dimensional concepts to one dimension. It has no particular relation to religion.

    yes. but, an important issue is whether you model a behavior as a continuous quantitative trait or a discrete meristic one with a few major morphs. religiosity and IQ are i think in the former category; many personality traits seem to be more in the latter. i state this because it looks like for personality we are finding quantitative trait loci of large effect (dopamine related), but for IQ we can not find these.

  • http://hanson.gmu.edu Robin Hanson

    razib, I think the core of your observation is that many, perhaps most, people just don’t put much of a premium on the coherence of their beliefs across different topics. It is jut not very important to them to have a coherent picture within which explicit beliefs and the implied beliefs which would make sense of their practices all fit well. I think this is true and helps us to understand many otherwise puzzling practices in many areas.

  • http://www.scienceblogs.com/gnxp razib

    robin, the general point is correct about what my point was. most human cognition is not explicit/conscious, and even explicit/conscious cognition is strongly shaped by implicit/unconscious variables. this is surely true of the nerds which populate these boards on a qualitative level, though i think quantitatively there is a difference insofar as OB readers probably pull a lot more of the weeds out of their mental garden on a regular basis than is the norm (part of this is probably due to there being less fertile soil in the first place, but part of it is a propensity toward this sort of self-criticism and training).

    on the specific point re: religion, it is important to model religion because so many people care so much about it (and will use it as an *excuse* for war or killing themselves, etc.). and, you have another problem insofar as most religious people today explicitly adhere to the idea that their views are coherent and systematic. in fact, they believe that there are contradictions between their own systematic coherent views and those of others, and that these differences are important enough to coalesce around to further collective action. not only will people claim that they have a coherent and systematic worldview, but, they imbue these presumed worldviews with ontological importance and are emotionally attached to them. in short, religion names normal cognitive biases and processes, and supercharges them with very powerful emotional valences.

    so, it is very easy i think for some atheists to confuse this ardor for something real which may be taken as an important variable in modeling human action. but, i think the cognitive psychology of religion suggests that these are shallow notional markers. e.g., most christians do not comprehend the athanasian creed to any substantive degree, despite disagreements predicated on its comprehension leading to violence and mass conflict.

  • no one

    razib: regarding The God Delusion, where do you think Dawkins discusses (and justifies) his understanding of religion? Which chapters (or page rangers)? I don’t remember much of this kind of discussion at all, and glancing back at the table of contents online to refresh my memory shows nothing that rings a bell.

  • http://www.scienceblogs.com/gnxp razib

    where do you think Dawkins discusses (and justifies) his understanding of religion?

    in the first half of the god delusion dawkins exhibits at least passing familiarity with the work of anthropologists such as pascal boyer and scott atran. but during the second half he pretty much ignores this and focuses specifically on abrahamic fundamentalism as the sine qua non of religion. my review of the book lays it out….

  • jeff borack

    “Our theory explains why simplicity is so highly desirable. To understand this there is no need for us to assume a ‘principle of economy of thought’ or anything of the kind. Simple statements, if knowledge is our object are to be prized more highly than less simple ones because they tell us more; because their empirical content is greater; and because they are better testable.” -Karl Popper

    The reason we want more testable models with greater empirical content is so that we can prove them wrong. If our model is wrong, and we don’t yet know how to fix it, what do we gain from continuing to use and build upon it? the statement “You’re wrong because your model is too simple. But I’m not going to tell you what your model is missing, at least not in a clear enough way to help you improve your model” is completely valid and useful in that it points out a problem. A solution would be nice too, but thats not always an option and not a necessary component of the scientific method.

    Do you think the reason-based choice bias effects our ability to accept a criticism of a model when a better solution isn’t included/available?

  • http://omniorthogonal.blogspot.com mtraven

    You do a lot better with religion if you don’t treat it as primarily a system of belief, but as a combination of a set of practices and a community that engages in them. Treat the beliefs as a secondary aspect and their inconsistancies and absurdities don’t matter so much. That is, if people are making sacrifices to propitiate the gods, it is the act of sacrifice, not whatever beliefs the practioners hold about the ontological status of their gods, that is the important feature. I’m not sure how this impacts the argument for or against simplicity, except that culturally embedded practices tend to be harder to describe (more complex) than abstract systems of belief.