Monthly Archives: October 2023

Biologically primary and biologically secondary knowledge; Merlin Donald and David Geary 2

Posted on by
2

In the previous post I summarised Merlin Donald’s model of the evolution of the human mind described in his Origins of the Modern Mind: Three Stages in the Evolution of Culture and Cognition. In this post, I look at David Geary’s model set out in his book The Origin of Mind: Evolution of Brain, Cognition, and General Intelligence.  As I said previously, the two books not only have almost identical titles, they also deal with very similar material. But the similarities are superficial.  Donald concludes that there have been three major transitions in human cognition, deriving his conclusions from artifacts indicating cultural shifts.  Geary bases his model of knowledge on a set of constructs, each of which is the subject of intense debate; motivation to control, intelligence, and brain modularity.

Motivation to control

The first construct Geary introduces is motivation to control (p.3) – a feature he sees as fundamental to all species in their struggle for survival.   Although he provides examples of what he means by the term (e.g. pp.72-80), I couldn’t find an explanation of which brain functions facilitate it, and the components of the construct remain rather fuzzy.  ‘Control’ could apply only to the immediate environment (e.g. food, water, shelter, social support), or extend to wanting world domination.  And I recall being required to write an essay on ‘motivation’ for an academic assignment – because it’s a great example of a term that’s used by different people to refer to very different things.  Ironically, motivation could be seen as a folk construct (of which more later); we all know roughly what the term refers to, but it’s so broad that to be used for research purposes it needs to be deconstructed.

Intelligence

Like Donald, Geary reviews the archaeological evidence showing the increase in relative brain size as the genus Homo evolved from its primate ancestors.  For Geary, the significant increase in EQ meant that H sapiens had superior general intelligence.  This enabled it to outcompete other species of the genus Homo for mates, food and other resources – which is how H sapiens achieved ecological dominance.   

Like motivation, intelligence is a term used to refer to different things by different people.  It’s been controversial since Charles Spearman came up with the idea of general intelligence (g) in 1904.  But that hasn’t stopped elaborations such as Raymond Cattell’s proposal that fluid intelligence (gF) is biologically determined, and crystallised intelligence (gC) is the outcome of an interaction between gF and sociocultural factors such as education.  Intelligence is, I’d suggest, another folk construct. 

Spearman defined intelligence in terms of particular academic skills (I’ve blogged about his model here). Donald (wisely) avoids using the term except in relation to Darwin’s views on animal intelligence.  Geary discusses intelligence and related brain function at length, but I struggled to pin down exactly what it was about the brain of H sapiens Geary believes resulted in their ecological dominance; I think it’s the development of the prefrontal cortex.  Geary draws attention to three levels of function in this brain area (p.211);

  • Monitoring and integrating information from posterior areas of the brain
  • Attentional control and inhibition of irrelevant information
  • Episodic memory and self-awareness.

The last of these three functions, Geary claims, gives humans what Endel Tulving called autonoetic awareness – enabling us to imagine ourselves in the past, present and future.  For Geary, this enables us to imagine a ‘perfect world’ and use problem-solving and motivation to control to try to achieve it (p.16).  He summarises (pp.304-5) his chapter on the evolution of intelligence in terms of Spearman’s g and Cattell’s gF and gC – again resorting to contested constructs.

Modularity

Geary’s model leans heavily on the concept of modules – areas of the brain that have evolved to process a specific type of information. Modules process information automatically and pre-consciously, in ways that during evolution increased the chances of an individual’s survival.  The automatic and pre-conscious processing has also resulted in inherent cognitive errors and biases.  The upshot is that we tend to configure our knowledge about the world in ways that aren’t always logical or rational; we default to folk biology (the natural world), folk physics (the way the natural world functions) and folk psychology (human behaviour and interactions).

The fact that some areas of the brain process some information automatically and pre-consciously (ie in modular fashion) isn’t in dispute – but the extent of the modularity is. Geary points out that the modules would have evolved due to environmental factors that were – at the macro level – invariant – but human beings also have to cope with a variant microenvironment.  He suggests the modules, although evolved for a particular purpose, are soft, (pp.11-122) ie they have some plasticity.  That’s very likely, but that characteristic by definition blurs the boundary between his categories of biologically primary and biologically secondary knowledge.

In addition to the motivation to control, intelligence, and evolved modules on which Geary’s model is founded, some other examples of him taking a construct for granted caught my attention; notably competition, the central executive, and folk biology.    

Competition

For Geary, evolution revolves around competition – initially social competition for mates, but in modern societies for occupational status (p.336).  Competition is certainly an important factor in the process of evolution, but the central feature of Darwin’s model was advantageous adaptation to the environment rather than competition as such; as Donald points out, competition usually arises only when resources are scarce.  And competition doesn’t explain the co-operation and altruism found in many human societies – a notoriously knotty problem for evolutionary psychologists.

Central executive

Another knotty problem is that of consciousness. I mentioned in the previous post that Donald concludes connectionist models of information processing indicate the central executive function (and therefore consciousness) isn’t modular – in a dedicated brain area – but is an emergent feature of a distributed network. 

Geary agrees with Donald – to an extent;  “… that novelty and conflict result in automatic attentional shifts and activation of the executive function is important because it addresses the homunculus question.   The central executive does not activate itself, but rather is automatically activated when heuristic-based processes are not sufficient for dealing with current information patterns or tasks…” (p.215).  But I couldn’t find any reference in Geary’s book to connectionist models despite their importance in cognitive neurology, and he appears to still see the central executive as modular. 

Folk biology

Geary makes frequent references to Charles Darwin and Alfred Wallace – 19th century contemporaries who each developed a theory of evolution.  I’ve referred above to Geary’s assumptions about the role of competition in Darwin’s model.

Another assumption Geary makes in relation to Darwin and Wallace (and Carl Linnaeus the taxonomist), was that their ideas must have been based on folk biology “driven by an interest in the natural world” (pp.188, 311).  Not only is this a somewhat tautological claim – if Geary’s model is right, all knowledge is ultimately built on folk knowledge – but he also overlooks the backgrounds of these eminent scientists.  Both of Darwin’s grandfathers (Erasmus Darwin and Josiah Wedgewood) were founder members of The Lunar Men, a discussion group whose members were leading scientists and industrialists, Darwin’s father was a respected doctor, Wallace’s father trained as a lawyer, Linnaeus’ father was a clergyman and amateur botanist, and Darwin, Wallace and Linnaeus attended grammar schools.  So from an early age, all three would have been exposed to far more than folk knowledge. 

What’s the difference between Donald’s and Geary’s models?

Key differences I noted were:

  • Donald sets out to create a coherent explanation of the evolution of human cognition.  Geary explores evidence that supports his model of human knowledge. 
  • Donald’s framework emerges from the archaeological, neurological and psychological evidence; the details of the changes are debatable, but the major shifts must have happened for the artifacts to exist.  In contrast, Geary tries to fit the evidence into a framework composed of broad – often contentious – constructs.
  • Donald dissects the heated debates associated with several constructs (e.g. modularity, laterality, speech and language, the central executive function, consciousness – but interestingly sidesteps intelligence).  Geary appears to take the constructs for granted. 

Conclusion

Neither of these books is an easy read.  And both necessarily involve some speculation because there are gaps in the archaeological evidence and in our knowledge about cognition. But Donald sets out his reasoning step-by-step so the diligent reader should end up with a good grasp of the evidence for the evolution of human cognition and the brain – even if his model is a little outdated because archaeology and cognitive neurology have moved on in the past 30 years.  Geary, in contrast, repeatedly tries to link up his constructs, which results in a fair bit of repetition, and left me struggling to see the forest through the trees. 

When I first read Geary’s The Origin of Mind my focus was inevitably on the factual information – which appeared pretty reliable. But I noticed he sidelined the debates about the implications of the factual information, and made assumptions about the constructs on which his model is based. My concern about Geary’s book is that teachers unfamiliar with cognitive neurology will be blinded by science (Donald has around 300 references, Geary cites well over 1000), but be unaware that Geary glosses over the reasons his key constructs are contentious, and that his model rests on assumptions.

Also, I couldn’t see the point of Geary’s model.  He assumes that by default students think in terms of folk biology, folk physics and folk psychology (biologically primary knowledge), so knowledge that’s not folk biology, physics or psychology (biologically secondary knowledge) needs to be actively taught.  It’s helpful for teachers to know that logical rational thought requires some effort because it’s swimming against the tide of the way human cognition works, but even Geary struggles to find a clear boundary between biologically primary and biologically secondary knowledge.  And teachers usually know what their students have learned with no apparent effort and what they’re having difficulty with.  So how does it help them to draw a somewhat questionable line between two types of knowledge? 

Lastly, despite Donald being an academic at a reputable university (Case Western), his book having a similar title, dealing with similar content, drawing conclusions about education, and being published by a reputable publisher (Harvard University Press) only a decade earlier, Geary doesn’t mention him.   I couldn’t help wondering why.

Biologically primary and biologically secondary knowledge: Merlin Donald and David Geary 1

Posted on by
Reply

David Geary is an evolutionary psychologist, whose theory about biologically primary and biologically secondary knowledge has been influential in some educational circles. 

The theory proposes that people acquire biologically primary knowledge with little effort due to the way our cognitive processes have evolved. Biologically primary knowledge is acquired using processes that are fast, frugal, simple and implicit.  It’s acquired naturally, in the course of development, and encompasses walking, talking, foraging for food, social interaction, and explanations about the world in the form of folk biology, folk physics and folk psychology.

Biologically secondary knowledge requires logical, rational thought – using processes that are slow, effortful, complex and explicit.  Unlike biologically primary knowledge, biologically secondary knowledge needs to be taught, and the main purpose of schools is to transmit biologically secondary knowledge.  I’ve previously critiqued Geary’s book The Origin of Mind: Evolution of Brain, Cognition, and General Intelligence here.

Geary’s theory pops up on Twitter from time to time, and during a recent discussion Oliver Caviglioli (@olicav) asked me what I thought of Merlin Donald’s work.  I wasn’t familiar with it, so read Donald’s Origins of the Modern Mind: Three Stages in the Evolution of Culture and Cognition.  You can see Caviglioli’s summary of the two books here.  I found the contrast between Donald’s and Geary’s books intriguing. 

One of these books is not like the other

The two books not only have almost identical titles, the content is also very similar; both open with a discussion of the archaeological evidence for the evolution of the hominin brain, discuss human cognition, and conclude with the implications for the transmission of knowledge.  But those similarities are quite superficial; in other respects Donald and Geary take very different approaches to their subject matter. 

Donald’s purpose is to provide a coherent account of how human cognition and culture evolved in tandem; in 1991, when his book was published, there were many competing explanations for human cognitive function, which he evaluates.  Geary in contrast, publishing over a decade later in 2006 when knowledge about cognitive function had made considerable progress, seeks to marshal evidence to support his hypothesis that there are two distinct types of knowledge.

Between them the authors refer to several issues that are the subject of heated debate, such as the extent of modularity and laterality in the brain, whether language is hard-wired or not, the nature of the central executive and consciousness, and the implications for pedagogy. Both authors come down on one side or the other of the debates.  Donald tackles the debates directly and systematically, taking the reader step-by-step through his reasoning.  I felt Geary glosses over the actual debates, and assumes the side that fits his model must be correct.  But first, a topic where both authors are in broad agreement – the archaeological evidence…

Brain evolution

Both authors open with an exploration of the archaeological evidence indicating how the modern human brain evolved.  Brains, of course, are composed of soft tissue, and decompose rapidly after death.  So the structure and function of early hominin brains has to be inferred from the size and internal shaping of the skull.  There’s a huge difference in both brain size and body size between early australopithecines (3-4m years ago) and Homo sapiens (from around 500 000 years ago).  To arrive at a fair comparison of brain sizes, researchers use an Encephalization Quotient (EQ) that denotes brain size in relation to body size.  Modern humans have an EQ which is almost 3 times that of our nearest non-human relative, the chimpanzee. 

Donald and Geary agree that the increase in EQ with the emergence of H sapiens was relatively rapid and relatively recent, and led to its ecological dominance; H sapiens proliferated, whilst other contemporaneous species (or sub-species) such as the Neanderthals and Denisovans died out, despite evidence that they interbred with H sapiens. 

The increase in EQ would have come about due to natural selection – genetic changes resulting in incremental changes in structure and function, that gave H sapiens advantages in interacting with the environment. Geary identifies three types of environmental factor; climatic, ecological, and social.  There’s no evidence of major climatic or ecological changes since H sapiens first appeared, so he concludes the critical environmental change must have been social.  Donald agrees about the climate and ecology, but proposes the significant change was a cultural one. Both authors are interested in what changes to brain structure resulted in H sapiens’ ecological dominance, but after that they part company; not over the archaeological or neurological evidence, but on how they evaluate the evidence and what conclusions they draw from it.  In this post, I’ll attempt to summarise Donald’s narrative.  In the next post I’ll look at Geary’s, and highlight the differences between them. 

Donald’s narrative; cultural change

Donald points out that evidence of the social organisation of hominins, and the artifacts they produced, are an indication of how the different species’ brains worked. The artifacts can be dated reasonably accurately, so we have a fair idea of the order in which they appeared, and which species produced them.  Donald traces back the chain of causality from the archaeological evidence to the brain function that would have been required to produce it. 

The brains of the earliest members of the genus Homo (such as australopithecus) appear to be similar to those of the great apes; their EQ was within the same range and, like the apes, they used tools.  The big shift came with the arrival of H erectus around 2m years ago.  Donald proposes that H erectus embodies the first of three key cultural changes that point to three key changes in the function of the hominin brain.  These are transitions from:

  • episodic to mimetic culture
  • mimetic to mythic culture, and
  • mythic to theoretic culture.  

Episodic culture

Many animal species appear to have episodic memory – the memory of particular events and associated sights, sounds, smells etc. Donald calls early hominin culture episodic culture – very similar to that of chimps.  Both had communities about the same size, could use tools, could mimic conspecifics using tools, and had a similar vocal apparatus – chimps in the wild use around 35 different vocalisations. 

Mimetic culture

For Donald, the first major cultural change was from episodic to mimetic culture.  By mimetic, Donald doesn’t mean mimicry; many species are able to copy behaviours.  Nor does he mean imitation (p.168) – copying the behaviour but adapting it.  By mimetic Donald means the ability to use representational symbols – mainly to communicate with others.  This ability opens the door to more sophisticated communication using vocalisation and gestures, to more complex social structures, and to advanced tool-making and hunting techniques; all characteristics of H erectus.  Not only did H erectus have a significantly larger EQ, but produced sophisticated tools, used fire, and migrated over long distances.

Mythic culture

The second major shift was to what Donald calls mythic culture.  What he means is the integration of fragmented mimetic knowledge into coherent wholes that provided shared explanatory narratives for communities.  The ability to integrate knowledge also resulted in the ability to reconfigure it, enabling problem solving.  So mythic culture involved innovation, which characterised H sapiens, who made novel tools such as needles, and created clothing and cave paintings.  H sapiens also showed significant changes in skull and jaw shape, suggesting the development of a vocal apparatus that enabled speech, as distinct from mere vocalisation – although there is evidence that speech might have evolved earlier.   A species that could communicate via complex speech and language would have an obvious evolutionary advantage over species that couldn’t.

Theoretic culture

Donald’s third cultural shift happened with the recording of symbolic representations on clay or wax tablets, papyrus, parchment, paper – or more recently electronic media.  Modern human cognition, regardless of the sophistication of brain function, is limited by working memory – we can process only a handful of items of information simultaneously, and only for a few seconds.  So regardless of how much knowledge we have as individuals or collectively, it’s difficult to recall mentally – individually or collectively – enough information to tackle very complex problems.  Writing (notably the use of an alphabet) changed all that.

Donald notes that Ancient Greece had an economy prosperous enough to allow some members of its society time to reflect on and expand existing knowledge.  And the cosmopolitan nature of Greek society prompted challenges to explanatory myths.  Both changes were facilitated by the development of a functional phonetic alphabet that enabled speech to be recorded, so people could read and critique earlier philosophical discourse.  Donald points out that accessing large quantities of written material required metalinguistic skills which in turn led to what he calls theoretic culture. 

Will brains evolve further?

The first two cultural shifts required an additional level of information processing in the brain; to enable the use of representational symbols in mimetic culture, and the integration of mimetic knowledge in mythic culture.   Donald makes the point that the development of a new brain function didn’t mean the functionality that evolved earlier was erased; he notes that in patients who have lost the ability to speak, the mimetic ability to communicate using representations such as gestures and symbols is often preserved. 

Donald’s third shift – to theoretic culture – required the development of an external memory store in the form of written material.  This has implications for brain evolution too. Brain structure changes with use, and individuals whose brain structure facilitates the use of external memory will be at an evolutionary advantage in a highly technological environment, which could result in long-term changes to the structure and function of the brain of H sapiens. 

Disputed concepts

Donald’s model is at some points necessarily speculative because the archaeological record isn’t exhaustive, and what happened during long periods of time remains unaccounted for.   But overall, his model is strongly evidence-based.  He explores the archaeological evidence, and how it’s been interpreted.  He explains how the modern human brain works – and what can be inferred when parts of it don’t work following injury. He tackles head-on, and explores in detail, some of the most disputed concepts in the field of human cognition, three of which I’ve summarised below; whether thought is mediated by language, the extent of modularity in the brain, and the nature of consciousness. 

Speech, language and thought

The development of speech and language was a pivotal point for the development of human cognition. Donald looks at: the lateralisation of the brain involved in handedness and language; how the vocal tract works to produce speech; theories about the extent to which language is hard-wired; and how speech and language are related to thought.  Contrary to much popular opinion, for Donald, evidence as diverse as that from aphasic patients and mathematical theorists show that internal symbolic representations precede linguistic representation.  

Modularity

In this context, a module is an area of the brain dedicated to processing a particular type of information, such as the visual information involved in face recognition, or auditory information involved in understanding speech. Modular information processing is automatic and fast, and introspection into the process is difficult (or impossible) for the person doing the processing. Donald concludes that some processes are entirely modular, others aren’t, and for some processes, it’s unclear. 

Consciousness

Consciousness is a phenomenon that has perplexed philosophers for centuries.  We’re unaware of most of the information our brains process, so how do we become aware of some of it at some times?  The function of integrating and evaluating information is often attributed to the central executive – although when the term was first used in Baddeley & Hitch’s model of working memory it wasn’t clear whether this function was modular (ie in a dedicated brain area) or distributed. 

The central executive introduces an infinite regress problem; if the central executive decides what information to pay attention to, what is it that’s paying attention? This is the same problem posed by the old idea of a homunculus in the brain directing operations.  Donald concludes that connectionist models of information processing suggest the strongest neural signals are synonymous with attention, indicating that the central executive isn’t modular, but is an emergent feature of the brain’s neural network.  That is, the central executive and consciousness are not distinct brain processes – they are outcomes of brain processes.

Implications for education

Understanding complex knowledge via written information requires metalinguistic skills that have implications for pedagogy.  Students not only have to learn a body of knowledge, but to access more knowledge they need to learn to read and write, learn the conventions of scholarship, and how to evaluate and critique arguments and theory.  In terms of current educational debate, just to be clear, Donald isn’t suggesting that ‘skills’ are more important than ‘knowledge’ – he sees both as essential.

Donald sees storing knowledge externally in books (and now computer systems) as essentially providing a new cognitive architecture that enables ready access to vast amounts of knowledge.   And – again just to be clear – he doesn’t mean that to access knowledge we can ‘just Google it’; his book was published in 1991, before the internet was widely available. 

In the next post, I’ll attempt to summarise Geary’s narrative, and to highlight the differences between his and Donald’s conclusions.