Evolution education: is our aim knowledge, understanding, or belief?

OLYMPUS DIGITAL CAMERANote: I have joined the “virtual class” component of Dan Kahan‘s Science of Science Communication course at Yale University. As part of this I am endeavoring to write a response paper in reaction to each week’s set of readings. I will post these responses here on my blog – my paper for week ten is below. Previous responses are here. I will also be participating in the discussion on Kahan’s own blog.

This week’s “questions to consider” (a reading list is here):

1. What is the relationship—empirically—between “accepting”/“believing in” the evolution and “understanding”/“comprehending” it?  Are they correlated with one another? Does the former have a causal impact on latter?  Vice versa?
2. What is the relationship—psychologically— between “accepting”/“believing in” the evolution and “understanding”/“comprehending” it? Is it possible to “comprehend” or “know” without “believing in” evolution?  Can someone “disbelieve” or “not accept” evolution and still use knowledge or comprehension of it to do something that knowledge of it requires?  Are there things that people are enable to do by “belief” or “disbelief in” evolution?  If the answer to the last two questions are both “yes,” can one person use knowledge or comprehension to do some things and disbelief to do others?  What does it mean to “believe in” or “disbelieve in” evolution? Is it correct to equate the mental operation or state of “believing” or “disbelieving” in evolution with the same mental state or operation that is involved in, say, “believing” or “disbelieving” that one is currently sitting in a chair?
3. What—normatively—is (should be) the aim of teaching evolution: “belief,” “knowledge,” or “both”?
4. If one treats attainment of “knowledge “or “comprehension” as the normative goal, how should science educators regard students’ “beliefs”?
5. If one treats attainment of “knowledge” or “comprehension” as the normative goal of science education, how should one regard political or cultural conflict over belief in evolution?

My response:

The empirical relationship between knowledge, understanding and belief

The evidence points strongly towards a distinction between knowledge and belief, for the simple reason that so many students have been able to demonstrate the former without the latter (or vice-versa):

  • In Hermann’s (2012) study, there was no statistical difference in understanding of evolution concepts between two extreme sub-groups of students, one believing in evolution and the big bang theory, and one not.
  • Sinatra 2003 (cited in Hermann) similarly suggests no relationship between belief and understanding of evolution
  • Blackwell found what could be termed a disjoint between understanding/application and belief: although an overwhelming majority of students selected appropriate answers when asked to categorize examples of specific evolutionary processes, the percent considering evolution the primary basis for progression of life on earth was 34-35 percent (with slightly different percentages for the two classes surveyed). The percent considering evolution compatible with their belief system was 27-29%, but only 6-9% said they could never believe in evolution.
  • Other studies found that it is common for students to believe in evolution without understanding it (Lord and Marino 1993, Bishop and Anderson 1990, Demastes-Southerland et al 1995, Jakobi 2010, all cited in Hermann).
A section from Blackwell's study, in which students had to categorize examples of various evolutionary processes.
A section from Blackwell’s study, in which students had to categorize examples of various evolutionary processes.

On the other hand, according to Hermann, several studies (Lawson and Worsnop 1992, Sinclair and Pendarvis 1997, Rutledge and Mitchell 2002) found that adherence to a religious belief system influenced the extent to which evolution was understood.

Defining our terms: the psychological relationship between knowledge, understanding and belief

The vagueness of the terms “belief,” “understanding” and “knowledge” obviously should give us pause when we are trying to make sense of these empirical findings.

I think we should try to define the terms in the way that is most fruitful to the problem at hand (while also seeking as much as possible not to create conflict with existing common usage, and to allow the above empirical findings to be applied). That problem is often put thusly, “Many students understand evolution, or demonstrate knowledge of evolution, without believing in it. Does this matter, and if so, what should we do about it?”

With this in mind we can come up with some rough working definitions:

  • Knowledge: Retained and retrievable true facts about physical properties and processes.
  • Understanding: A deeper form of knowledge accomplished by forming multiple connections between true facts on the subject at hand (evolution) and between the subject and others. (Similar to Gauld 2001’s definition, cited by Smith 2004.) An example of one such connection may be between a scientific theory and its supporting evidence (as suggested by Shipman 2002, cited by Hermann).
  • Belief: A committed, often emotional attachment to a proposition, which itself may be true or untrue, falsifiable or not. I would argue that it is sensible to talk both of faith-based belief in religious precepts, and of belief in scientific theories, which can be driven by thorough understanding of their scientific basis, or by blind faith in scientists. (I subscribe to the summary by Gess-Newsome 1999 [cited by Smith], of knowledge as “evidential, dynamic, emotionally-neutral”, and belief as “both evidential and non-evidential, static, emotionally-bound.”)

These definitions mesh well, I believe, with most of the empirical findings we read about this week, including Hermann, Smith and Everhart (2013). Hermann, for example, builds on Cobern’s partitioning concept to conclude that religious students view science ideas as “facts,” categorized differently than beliefs, which have a stronger emotional attachment. This helps students compartmentalize because they have created an “emotional distance” between scientific conceptions and religious beliefs.

In creating these definitions I have had to dismiss definitions that I think are unhelpful for the problem at hand. For example, according to Hermann, Cobern (1996) stated that knowing “is the metaphysical process by which one accepts a comphrehended concept as true or valid.” But this definition is actually much more like belief, as most of this week’s reading understands it.

I’ve also had to discard the philosophical convention that belief is a necessary condition of knowledge (Smith). When describing the way that people learn, and knowledge acquisition’s interaction with existing belief systems, this stipulation just doesn’t make sense (given the evidence we have of knowledge without belief). By casting off the impractical philosophical definition, I resolve a problem that Smith recognized – that if knowledge is dependent on belief, science education must foster belief.

There will always, I think, be messy edges and overlap between these realms. For example, it is hard to think of much useful knowledge that we can retain as utterly isolated “facts.” Facts that are part of a coherent schema are easier to retain or retrieve. We do, however, remember groups of facts that are connected in greater or lesser degree, both to each other and to other facts and schema in our brains. The difference between knowledge and understanding is thus one of degree.

Is lack of belief a problem? Or is it lack of understanding?

It should be noted that the issue with religious students’ non-belief in evolution is not merely one of semantics or a confusion of terms. The problem is we are not satisfied with students merely believing evolution in the way that they believe in discredited Lamarckian or Ptolomeic ideas. We don’t want them simply to believe “that evolution says x”: that implies that evolution has no special empirical status and it may as well be false, as those outdated scientific theories are. A student who can say only “that evolution says x” is merely parroting scientific language. She is in truth only a historian of science rather than truly a scientist herself – and I think that’s what so bothers us about the learning outcomes exhibited by students like Aidan and Krista, in Hermann’s study. We come away with the sense that their knowledge falls short of true scientific understanding.

I agree with Smith, however, that we should not go so far as to seek or require belief – or perhaps, I might say, “complete belief.” It is not and should not be the goal of a science class to completely overhaul students’ worldview, religion and all.

What we are seeking is for students to believe something like:

“Evolution, which says x, is the best supported scientific way of understanding the origins of various species, the way species adapt to their environment, etc etc.” (A conclusion similar to Smith 2004.)

And this requires an understanding of evolution, in the strong sense of understanding, which encompasses comprehension of justification. One may even argue that this type of belief follows necessarily from strong understanding: that is, if you understand mechanism of and scientific basis for evolution, and the comparative paucity of scientific explanation for other theories of species’ origins, then you will necessarily believe that “Evolution, which says x, is the best supported… etc, etc.” This could be a neat logical maneuver to employ because it means that we can avoid talking about the need for students to “believe in” evolution – which carries a lot of nasty cultural baggage – and just talk about understanding instead.

While several empirical studies have demonstrated that students can easily demonstrate knowledge of evolution without belief in evolution, understanding is a much more slippery eel. As previously alluded to, understanding encompasses a wide spectrum, starting from a state barely stronger than plain knowledge. But I would argue that understanding evolution, in its strong form, encompasses an understanding of the scientific justification for the theory of evolution – and that necessitates an understanding of the nature of science (NOS) itself.

Nature of science: the path to strong understanding of evolution

The best tactic for accomplishing this right kind of evolution belief, or strong understanding – and happily, a key to solving much else that is wrong with science education today – is to place much more emphasis on the scientific method and the epistemology of science. This includes addressing what sorts of questions can be addressed by science, and what can’t; and also the skeptical, doubtful tension within science, in which things are rarely “proven” yet are for good reason “believed.” Crucially this involves helping students to understand the true meaning of “scientific theory,” whose misunderstanding often underpins further misconceptions about evolution’s truth status.

This effort also involves exploring the tension between self discovery and reliance on authority – acknowledging that it is important for students to learn to operate and think like scientists, and we want as much as possible for them to acquire knowledge in this way: but that the world is far too complex for us all to gather our own data on everything. So students must learn how to judge the studies and reasoning of others, how to determine what evidence from others is applicable to what conclusions or situations, and how to judge who is a credible expert.

Misunderstandings of the nature of science (as well as certain broad scientific concepts) often lie at the heart of disbelief in evolution, as Hermann illustrates. In his qualitative study, both students showed a poor understanding of the methods and underlying philosophy of science, displaying a need for truth and proof – despite their good science knowledge performance.

Smith, rather inadvertently, gave another example of this problem. He cites a student who wrote to Good (2001):

I have to disagree with the answers I wrote on the exam. I do not believe that some millions of years ago, a bunch of stuff blew up and from all that disorder we got this beautiful and perfect system we know as our universe… To say that the universe “just happened” or “evolved” requires more faith than to believe that God is behind the complex organization of our solar system…”

Good uses this passage to justify making belief a goal of science education. Smith takes a contrary view, that “meaningful learning has indeed occurred when our four criteria of understanding outlined above have been achieved – even if belief does not follow” (emphasis in original). Instead I would argue that the student does not understand evolution in a meaningful way, having false impressions of underlying scientific and philosophical concepts such as entropy, order, and Occam’s razor.

Will nature-of-science education work with all students?

The research outlined above shows a mixed prognosis for our ability to overcome these issues and foster belief in the evolution proposition. Everhart’s work with Muslim doctors suggests that most participants considered subtly different meanings of the theory of evolution, and could consider evolution in relation to different contexts, such as religion and practical applications, with attitudes to evolution changing when the relative weights of these meanings were shifted. These meanings include a professional evaluation of the theory that could be held distinct from other evaluations. This suggests that participants may recognize the truth of evolution within a science epistemology framework, which should be sufficient for belief in our proposition, and not give evolution the same status within other, more personal epistemologies.

But Hermann suggests that students ultimately fail in integrating science and religion, which creates a fear of losing religious faith, causing the student to cling to the religious view while further compartmentalizing science concepts. This drives at the hard, hard problem at hand: even with a perfect understanding both of evolution and of the nature of science, religious students are likely to run into areas of conflict that create psychological discomfort. This is because the epistemic boundaries of science and religion are neither hard nor perfect. Some of the areas that science claims as well within its remit to explain – such as the age of the earth – run into competing claims from religion.

One way out of this conundrum is for a student to redraw the boundaries – to say, OK, I accept the scientific method where it does not conflict with my faith; but on this matter I must reject it. Hermann’s subjects appear to have done this to a certain extent, but run up against limits. I would hypothesize that this line-drawing process itself leads to further discomfort, especially among students who are brighter and/or show greater understanding of the nature of science, because they would consciously or unconsciously recognize the arbitrary nature of line-drawing. And unfortunately, one good way to resolve that discomfort would then be to discredit the scientific method.


8 Replies to “Evolution education: is our aim knowledge, understanding, or belief?”

  1. I’d be interested in your reaction to two points I raised when discussing these questions over at Dan’s site.

    1. What do you think of the example of classical Newtonian mechanics, which physicists routinely use in their work as if it was absolutely true, but which they actually know to be false (because of things like quantum mechanics and relativity)?

    2. What do you think of the idea of scientific beliefs/knowledge as “models of reality”?

    A model is a means of making predictions about what will happen or be observed in controlled circumstances. Physical theories like Newton’s laws of motion are often regarded as a type of model. But models can be multiple, inconsistent, overlapping, and of greater or lesser fidelity. A scientist can use a model even knowing it is wrong, and then switch to using a completely different model inconsistent with the first, without feeling the slightest twitch of discomfort. As George Box said: “All models are wrong, but some are useful.” Does usefulness substitute for truth?

    If a model of reality has been shown to be useful at making predictions, does that constitute “knowledge”?

    1. Hi NiV,

      1. I think it’s not quite right to say classical Newtonian mechanics is false. It is true for the purposes of physics and engineering on our ordinary human scale (ie neither a quantum scale nor an relativistic one). So to a certain extent you can operate as if it’s true, and I’m not sure there’s any contradiction there. In any case, on first reflection I don’t see the relevance to our discussion, because there is no similar model in evolution – ie there isn’t a model that works fine on one scale, and then turns out to be completely incorrect on a different scale. But…

      2. That’s an interesting point. I think it’s very important to talk of “models of reality” because that shows how flexible and variable scientific conceptions of the world are, even conceptions held by experts. If we think of “model” in a psychological, reductionist sense as “all the interrelated concepts that make up ‘evolution’ in the mind of the individual, and which she will use to frame further thinking about evolution,” the model of evolution held by a microbiologist will be different from that held, says, by a botanist. One model might describe certain concepts, less crucial for that individual, on a broader scale that is perhaps slightly flawed, but works well enough for the task at hand.

      Therefore it would be reasonable to conclude that an absolute model of evolution that is true in all circumstances is probably an impossibility.

      But I wonder if all this is a bit beside the point – after all, when we’re talking about high school students, we’re not looking to get anywhere close to that impossible standard. We just want them to understand evolution well enough that they can comprehend the origin of species, selective breeding, bacterial resistance to antibiotics, etc. So yes, in most cases I’d say understanding of a useful scientific model = knowledge.

      1. Thanks. That’s interesting.

        “I think it’s not quite right to say classical Newtonian mechanics is false.”

        How so? Is it true or false? Or can it be both, depending on the circumstances?

        “It is true for the purposes of physics and engineering on our ordinary human scale (ie neither a quantum scale nor an relativistic one).”

        Isn’t the human scale a quantum/relativistic scale too? The predictions of quantum mechanics and relativity still *work* at the human scale.

        If I understand what you’re saying, it is that within the specific frame of ‘human-scale physics’, that Newtonian physics *is* true. Truth is frame-dependent, and can differ in different frames. The world is divided into many boxes, and each box has its own definitions of true and false inside it.

        What you say sounds like an interesting viewpoint on “truth”, and feels psychologically valid to me. But is it what you meant, or am I misunderstanding?

        “I don’t see the relevance to our discussion, because there is no similar model in evolution – ie there isn’t a model that works fine on one scale, and then turns out to be completely incorrect on a different scale.”

        Oh, but there are!

        You can go a long way in understanding evolutionary effects by thinking of it in terms of the features of organisms being part of a “design” to achieve specific aims. A virus has surface proteins “for” breaking into cells. Actually, no. Viruses have surface proteins, and these have the effect of breaking into cells which result in the viruses with those proteins multiplying. The “design” frame recruits our intuition about intentional designs to predict what features we ought to expect organisms to have or not have, and evolutionary biologists use the mental model all the time. The trick is to recognise the limits beyond which it doesn’t work, and switch to a more accurate but more difficult-to-calculate-with model.

        Likewise, you can go a long way by assuming the aim is for individual organisms to survive, (or even the species). The model works a lot of the time. It’s only when the interests of the genes conflict with those of the organism that the model doesn’t work, which is why altruism used to be regarded as a problem for evolutionary theory. “Survival of the species” is a model that works on a broad scale, but not when you look at the finer detail.

        Models appear everywhere!

        “But I wonder if all this is a bit beside the point – after all, when we’re talking about high school students,”

        I think models are how all humans think. Scientists analyse it more than other people, and formal reasoning makes the transitions more visible, but high school students surely do it too.

        I suspect it is the failure to point it out, to give students the impression that scientific truth is absolute, and not relative to a particular model of the world, that causes the confusion. They end up confusing the models scientists use for truth, and then can be convinced against it when the flaws in the models are pointed out. Or on the other side, retain their belief that a flawed model is absolutely true, and consider the contrarians to be dishonest or stupid for disagreeing with it.

  2. Knowledge: understanding a statement S.
    Belief: assigning a probability p that S is true.

    So a student can have knowledge of evolution, by understanding what it is, but not believe in it, by assigning a low p.

    1. Hi Michael – I assume you don’t mean “assigning a probability” in any conscious sense, right? I just don’t think belief works like that, most of the time.

      But whether the “strength of our conviction” basically amounts to a subconscious probability – hmmm, that’s an interesting idea!

  3. Tarmar wrote
    What we are seeking is for students to believe something like:

    “Evolution, which says x, is the best supported scientific way of understanding the origins of various species, the way species adapt to their environment, etc etc.”

    Thinking from a petagogical perspective (and I’m not sure what perspective you are taking on this) it is sufficient to have students understand the theory and something of the “scientific way of understanding” that supports it.
    If the goal is to impart some belief in the philosophy of science or to understand the TRUE meaning of “scientific theory” those goals can be better accomplished using other topics as illustrations. Anything more than this type of mimimalist approach suggests to me that unacknowledge assumptions or beliefs are coloring the conclusion.

    FYI, I also think you jump too quickly to base your reasoning on the philosophy of science especially when attempting to dis-entangle concepts of belief (and therefor probably philosophy too). I would expect someone more fully appreciative of the issues and disagreements in the philosophy of science would make some acknowledgement of the of the difficulties being passed over in references to “understanding of the nature of science”.

    It seems that one of the few key points about the philophophy of science that philophers agree on is that most of the rest of us, including most scientists, often appear to them to hold naive notions regarding the philosophy of science.

    I’ve enjoying and followed your postings over the course of Dan’s class.

    1. Hi Cortlandt,

      Not being an academic, I take whatever theoretical perspective captures my fancy, which could be both a good and a bad thing. 🙂 I think this essay was basically mashing up psychology and philosophy in a very undisciplined way, to reflect mostly on pedagogical research.

      On your first point, I would say I proposed that target belief not just because I want students to understand the nature of science or of theories – although that is a concern. What I was trying to get at is that if one truly understands the rationale for the theory of evolution, then one will necessarily believe it to be the best supported theory for understanding these things. So it’s really just a way of getting at a deep understanding of evolution.

      But I think you’re very right that the polarized subject of evolution is not the best place to try and train students in scientific thinking. Probably better to develop that over a good half dozen other topics, and then treat evolution as just another example of the scientific method at work!

      Yes, I’m sure there are a lot of philosophical disagreements that I glossed over for lack of background/understanding. Would love to read more about the philosophical debate over what constitutes “understanding of the nature of science.”

      Thanks for your comments!

  4. RE: the philosophical debate over what constitutes “understanding of the nature of science.”

    http://plato.stanford.edu/entries/pseudo-science/ has a good primer on “the demarcation problem” which is one entry point into the larger question.

    I’m think we must understand the “nature of science” from several perspectives. One view of the nature of science is that it’s quite often second rate. For instance I recently discussed a paper on culturalcognition.net that IMO should be recognized as “poor science” because the paper did not publish key survey results. In my mind this violates the scientific method. In a world where science writers were concerned with high quality science then their reporting of the paper would make note of the shortcoming. I think should be normal for science journalists to identify the paper in question and the science journal that published it as “second rate” or words to that effect.

    I think it’s a good exercise for students of critical thinking to have an appreciation of the difference between
    “astrology has been disproven by science” and
    “astrology is not supported by science”.
    The latter meaning that tests of astrological predictions fail to refute the null hypothesis — the predictions do no better than chance.

    I like the question “Is astrology scientific”?
    I suggest that a “true” understanding of science means that one can defend an answer of “Yes, it’s scientific”. Indeed, thinking about how to design an experiment of astrology is an interesting learning exercise. And, by the way, such experients have been run and to my knowledge it does no better than chance; in layman’s terms astrology has been scientifically tested and it fails.

    If you say “No, astrology is not scientific” then you need to explain why it does (or doesn’t) make sense to say that astrology has been tested by scientific experiments … and “fails”. Isn’t the experimental test itself scientific? Because a theory ‘fails’ a experimental test does that make the theory non-scientific? This example points to the several meanings of “scientific”.

    To anticipate one objection, yes, most practicing astrologers probably pay little to no attention to the scientific method. But at the same time I’m not paying attention to the massive amount of experimental science and software logic which enables me to write this comment on my computer and send it to you.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s