Hi everyone
So who saw the x-files on wednesday? Mulder raised the point that perhaps
evolution is not really a process of steps. Instead he referred to
punctualism, or punctual equilibrium, where evolutionary advances are
Hi,
In Dawkin's "The Blind Watchmaker" he talks about punctualism (I can't
remember which chapter) and argues against it quite convincingly.
Unfortunately I can't remember much of his argument but I think he said
something about missing fossil records giving the impression of
fits and*
starts whereas the reality is much smoother; also that the statistical
likelihood of enough mutations happening at once to
suddenly do something like create an eye is small enough to be
impossible if you consider the slow mutation rate. $
*Correct - see also my previous posting on Gould etc.
$ One point here is that a small genetic change can have a large phenotypic effect - this is Goodwin's point about numbers of fingers etc. In a lot of cases (development but also behavioural decision-making) a continuous underlying distribution is quantized. Eg the number of fingers depends on the amount of precursor cartilage in the limb bud. If there is a slow genetic change to increase / decrease the amount there will be an abrupt and apparently major change in finger number when the threshold is passed. This looks like (and from the view of the phenotype actually is) a Mulder-type jump but the actual genetic control is perfectly 'respectable' slow change. So do you look at the phenotype (which is what is actually being selected) and say 'jump' or the genotype and say 'gradual evolution'?
A behavioural example (Byrne 'Thinking Ape' has details, but they will be elsewhere too). A lot of primate foraging involves looking for signs of buried bulbs; like animal tracks they don't look like what you are looking for, and they tell you that something was there in the past. Formally this is similar to writing which refers to things which aren't present, and looks nothing like the things if refers to. If skill in 'tracking' which evolved slowly in the primates is combined with skill in making things (which evolved slowly in apes) to make a combined skill in making and tracking the meaning of your own signs the result is an abrupt transition to much greater possibilities in storing memes for future reference.
Going back to structure; usually (particularly if things are lost) the genes for them remain, as there is little selective pressure to remove them now they have no phenotypic effect. Thus chickens sometimes develop teeth and by adding to the precursor cartilage it is possible to get them to develop bones which have been lost in modern chickens. There is therefore 'reserve' flexibility, though this doesn't apply to completely novel developments (eg it's unlikely that the ancestors of birds had 'hidden' feather genes).
Parallel to this, animals usually have a range of alternative behaviours,
dependent on the environment. For example, Zahavi's detailed studies of HATN
show that helpers change tactics according to circumstances. In general helping
seems to be a transition depending on how the environment (availability of
territories, food etc) fits the helping threshold. For example the endangered
As I said during the lecture much of the more detailed material on the ways that genes affect development is not critical if your primary interest is in the topics of week 4 on and you are happy to take the 'gene for' as a black box. However if you want to criticise the possibility of genetic control over behaviour you will need to show awareness of how genetic control operates.
Notice that openness to environmental influence shows that the genes offer a wide range of alternatives to select from (and that this range may not be exactly specified - e.g. human manipulative skills work with ANY objects of about the right size) but; resistance to environmental insults implies fairly close genetic control.
You need to know, as they will appear again;
the difference between proximate and ultimate causes (how to they relates to Tinbergen's 4 whys?)
what the EEA and/or assumed environment are.
Can you relate these to the incestuous sibs?
With reference to the essay question discussing the occurrence of the
behavioural development of vehicles (as influenced by replicators) I am
slightly unsure as to the role of neuroembyrology.
If 'bootstrapping' suggests early stages of development provide the basis
for the animals reactivity to the environment thus influencing genetics
later on, Is this the same as Oppenheims Ontogenetic Adaptation Theory
(1984)? (That early phases are preparations for adult features and the
extent to which they are adaptations to current conditions).
Don't know this one, but see below.
Sorry if this sounds rather jumbled but I'm totally confused!!
So is development!
The development of any organism has to meet several constraints;
It must function and meet selection pressures at all stages; It must be as economical as possible in both genetic and environmental input (the KISS principle - what isn't there can't go wrong);
It must be able to compensate as far as possible for environmental interference, so that normal development is maintained even under abnormal conditions;
It must have evolved from ancestors which were different, but constraints 1, 2 & 3 operate at all stages in the evolution.
For example, you have to function adequately as a foetus, a newborn and an adult, but evolved seamlessly via reptile, amphibian and fish ancestors so:
placenta etc. derive from structures evolved to feed the embryo in the ancestral egg;
the development and design of your larynx and associated nerves and blood-vessels derives from the developmental process which produced gills. I think I talked about this in the lecture? In this case developing a new set of genetic instructions would have risked failure under 4, so the muscle-blocks for the gills appear first (as part of general segmentation). They provide an 'environment' which acts as a boot-strap for cartilage, nerves and blood-vessels; once these are in place they provide an environment for the development of the correct larynx structure. Doing it this way means that the later stages can compensate (within reason) for any irregularities of the earlier stages.
Similarly the mouse uses its whiskers to boot-strap development of the corresponding brain areas and the kitten boot-straps visual cortex development from visual experience. In both cases development automatically fits itself to the body size of the individual and the local environment. (If the development of binocular vision was directly controlled by the genes, there would have to be a link between the development of head size and the brain mechanisms co-ordinating the eyes - this would be far more likely to break down than the boot-strapping mechanisms actually used which check and adjust).
Imprinting in ducklings etc. also allows better camouflage etc. to evolve as the genes instruct 'learn what you see first'. If recognition was genetic, both form and recognition would have to evolve together and there would have to be a genetic link between the two. There are parallels with human language learning - fairly general constraints and 'learn what you hear' allow the rapid evolution of languages.
Language also brings up the infant/adult aspect of development. Structures/abilities which are of use to the infant should be lost if possible if they cost - thus you lose the ability to learn languages easily after childhood and have to rely on general-purpose language learning. However getting rid of your navel is more trouble than it's worth despite its lack of function. Adult form may therefore keep some characteristics which are of no use to the adult.
A second possibility is that juvenile patterns may be recycled for adult use in a different context - this is an economical way of keeping mechanisms simple. A wide range of species use infantile patterns as affiliative / non-aggressive behaviours e.g. to show submission, to show affection, and in courtship.
Juveniles may show adult patterns (often in play) - especially in species with relatively short lives where it is essential to be fully skilled as an adult immediately. Play fighting and courtship are the most obvious but hunting, nest-building and parental behaviour also appear. Again it is most economical to boot-strap e.g. by young birds fiddling with sticks while they are still in the nest and developing the ability to handle them by experience rather than having all the information about sticks genetically coded in.
NOTE The constraints of 4 don't apply so strongly to behaviour because it does not have to be shown all the time. Thus human structure clearly shows our origin from fish - human behaviour doesn't.
After considering the references from week 2 concerning behaviour and
genetics, I am not sure exactly how an answer on this topic may be
tackled. It seems that much of what I have read covers development of
embryo form, which indirectly will lead to certain behaviour. Is this a
valid interpretation, and is this what an exam question in this area would
be seeking knowledge of?
Yes
1) The mantra 'behaviour is structure in action' What is actually being selected is behaviour, and there is evidence that the effect of behavioural experience is to modify structure and in turn to modify later behaviour - i.e. a developmental tuning process occurs.
2) Purely from a practical point of view, bodily structures are easier to study than CNS structures. We could look at;
Bodily development (Edelman etc.)
CNS development (Edelman etc.)
Development of behaviour through maturation (e.g. improvement of pecking in chicks with age, Bower et al on development of reaching skills in babies, early linguistic development). Here change in performance over time is what is actually being measured; we may have the links back to CNS development (less for babies, where we have the best behavioural data, for ethical reasons).
Note a problem which is particularly salient for behaviour;
if the system is 'designed' to develop emergently using feedback from experience, testing alters experience feedback CNS development later behaviour, making it difficult to assess how the system works in longitudinal studies. Cross-sectional studies have the problem of discounting individual variation.
The strongest evidence for genetic control and maturation is where the system is NOT responsive to environmental influence;
e.g. for speech, neologisms such as 'wented', children who invent a creole which is better than their parents' pidgin;
deterioration of reaching with age (and experience) as the baby moves from using one CNS system to another (Bower).
Conclusion; your best evidence would be on devt changes in behaviour but given that the complete path from genes to behaviour is usually unknown, you may have to mix examples where behaviour is known and structure isn't with examples of structure 'without' behaviour. (Note that, as I said, even a mollusc shell or a bone 'behaves', much as a tree does).
CHANGE
DURING THE LIFE-CYCLE?
I just wanted to check that i am right in thinking that animals do not
actually change their features during life as an adaptation to survive the
environment and competition. It is just that those who have better genes
(and so better phenotypes) survive and reproduce.
Depends what you mean by 'features' -
For physical structure - most animals are capable of the sort of change human adults can achieve (increased fitness from excercise etc.) but there are a few (like serially hermaphrodite fish which change can from female to male if the resident male dies) which can achieve more dramatic change - but these are atypical. During development, structure (including brain structure) is much more dependent on environment - e.g. the stuff about experience and visual development (see Michel & Moore) I talked about in week 2. There are also dramatic effects of language exposure, sib position etc. in humans and effects of hormonal influences across a wide range of species during development.
If you mean behaviour, all aspects of learning are relevant - I underplay these as they are dealt with in other courses (and you can use material from other courses) and ethology has traditionally stressed the fixity of behaviour, not the lability.
It's worth remembering that adaptability has to be genetically specified, e.g. if you compare humans to chimps:-
both species have two eyes (genetically specified and little influenced by environment);
both have binocular vision (development requires environmental input to interlock with the genetic specification, if development is to be normal);
chinps are genetically unable to use a speech-rich environment to develop speech - only human children have the genetic ability to decode and learn from speech they hear.
Chimps' reaction to speech is like our reaction to AIDS - no genetic ability to respond in a meaningful manner.
Also in apes is grooming a conscious strategy?
Yes - see Byrne, Dunbar or de Waal on the reference list. The evidence is
that apes (and monkeys) groom those who would be of most use to them.
CRITICAL PERIODS
Apart from the whole controversial issue of 'design', I was wondering whether
you had some ideas and/or references for the evolutionary advantages of
plasticity and critical periods. Whilst the former is more obvious, I'm having
some trouble in finding evidence for the adaptive value of critical periods.
Try Steven (?) Pinker's 'The Language Instinct' and / or 'How the Mind Works'
The two basic reasons for critical periods are
1) learning costs in terms of energy - e.g. the metabolic rate of the brain rises 50% while language is being learnt (hence Ridley 'The Red Queen' the emphasis on wit in courtship as it shows that you were well-nourished at the critical period of early speech development). Once something (e.g. speech) has been learnt there's no point in pating these costs any more. This is especially true in technologically stable traditional societies e.g. I'm reading a couple of books on warship design which show there was more change between 1860 and 1880 than between 1560 and 1860 - pre-1860 people mostly didn't need to relearn skills in their adult lifetime.
2) continuing learning risks over-writing useful learning with subsequent information e.g. one might over-write memories of harsh weather conditions and how to deal with them with memories of more recent milder seasons - as I said on the course the inability of the old to learn means they stay as an 'archive' for their younger descendants.
Incidentally Dennett (Consciousness Explained) I think discusses why instinct is better than learning - it ensures you get the right answer first time - learning is only better when the environment is predictably unpredictable. People tend to overvalue learning but they have only been the dominant group for 30,000 years - most dinosaur species lasted 5-15 million years and the group 200 million years!
EVOLUTION
AND DEVELOPMENT
> I KEEP GETTING CONFUSED AS TO WHAT THE QUESTION ON EVOLUTION AND
> DEVELOPMENT IS ASKING EXACTLY. IS IT STATING THAT DEVELOPMENTAL AND
> EVOLUTIONARY CHANGE ARE GRADUAL AND THEN ASKING WHETHER EVOLUTION AND
> DEVELOPMENT ARE LINKED?;
This would be the most likely approach.
OR IS IT ASKING US TO DISPUTE THE FACT THAT THEY
> ARE BOTH GRADUAL AND THAT INSTEAD EITHER ONE, NEITHER OR BOTH MAY ACTUALLY
> INVOLVE PUNCTUATED EQUILIBRIUM OR SOME OTHER ALTERNATIVE?
This would be a valid alternative; as I said in week 1/2 and also in week
10, relatively slight changes in development (legs and language) may have
extremely far-reaching effects so gradual change in ontogeny (development)
may lead to major change in phylogeny (evolution).
>
> PLEASE COULD YOU GIVE ME SOME IDEA ABOUT WHICH ASPECT TO CONCENTRATE ON AS
> I DON'T WANT TO GO COMPLETELY OFF THE RAILS!
The essay questions I set are intended to allow a range of acceptable
answers - there is no single 'right answer'. See also my other posting on
this question which I have sent to the ng. Provided you have got the basic
concepts correct (as you seem to have) and can produce a coherent argument
you should be OK. If you're in doubt, send me an outline or draft.
>
> THANKS
>
Sean
Goodwin reports to us in 'How the leopard changed its spots'(p35) that
John
Lamarck's theory of adaptive inheritance.
I was just wondering if any further evidence has arisen to support
et al's previous findings or if Weismann's barrier is still firmly in
place.
I haven't heard anything to indicate that Weismann's barrier* has been seriously challenged - certainly at the level of complex animals with 'real' behaviour, but I'm not fully in touch with the biochemical work.
*Weismann's barrier is that changes in the body resulting from experience don't get passed down to descendants - so if you get your ears pierced and learn about Wesimann's barrier, your baby will be born with unpierced ears and knowing nothing about Weismann. If this barrier did not exist, Lamarckism (the inheritance of acquired characteristics) would be possible.
It's worth bearing in mind that replicators form a larger proportion of simpler organisms, and that they are less buffered by the vehicle from their environments.
E.g. in a virus, or Cairns-Smith's clay and organic replicators (covered by Dawkins) there is no vehicle and the replicator gets the materials to replicate directly by affecting its environment (e.g. viruses which infect people reorganise the cell to make it make more viruses).
What follows is my speculation.
Suppose the cell/environment is deficient in the usual materials but the virus ends up functional despite being deficient as a result of the shortage i.e. it is missing some units and has therefore mutated. In this case the mutation is due directly to the environmental change.
As I recall
As a student I was taught about homeostasis - i.e. the organism regulating itself despite changes in the environment - homeiothermy (warm-bloodedness) is an obvious case. The point is that most mutations due to changes in the environment will, like all mutations, be handicaps. Organisms will therefore try to avoid them - and they will evolve to make their vehicles buffer their replicators from environmental disturbance.
My guess therefore is that Weissmann's barrier is an expression of the effect that well-adapted organisms avoid being disrupted by the environment if they can; those which can't are selected against. Primitive organisms are less good at it but most can afford to take the losses as they have (?have to have/ have only survived because they have) high reproduction rates.
1) Thanks for your previous reply; can you explain what HATN is - I'm sure
I'm being very dense/ignorant but I can't decipher it!
Helpers At The Nest!
2) I was reading Gazzaniga's 'Nature's mind' - in the intro there seems to
be some confusion about the term 'selection': on the one hand he seems to
be saying that it refers to natural selection - in this case the selection
of complex neural structures in as much as they correspond to animals'
(humans') behaviours.
Correct - but see below. The problem arises because; there don't seem to be any good alternative words to selection; the authors are trying to make the point that the same process is occurring under different time-scales - the genes provide alternatives; the one which best fits the environment is selected. In natural selection the 'duds' are destroyed with the body containing them - in developmental selection they merely become 'excess baggage'. One reason for 'instincts' developing is that providing alternatives has costs - if they are never used these costs are unnecessary (the 'Titanic' argument - why have lifeboats you will never need). In the short term (evolutionarily) instinct is selectively advantageous - in the long term some selective iceberg always turns up which is why there are so many extinct species.
On the other, he seems to be saying that it might
refer to a selection process within the individual - when an individual
encounters an environmental challenge, it selects the most appropriate
behaviour from the repertoire available.
Correct - but a major point is how 'prefixed' the repertoire is. The extremes could be;
'classical' Lorenzian instinct where the repertoire is prefixed to only one possibility;
'classical' Pavlovian/Skinnerian conditioning where the repertoire is infinite;
intermediates!
By this definition, selectionism
would appear to be a far more radical approach, since he seems on the
verge of denying any environmental influences, all learning/development
being simply a case of 'remembering'/discovering the appropriate
behaviour.
The critical question is how much of the information is 'pre-remembered' and just has to be 'discovered'.
He also seems to be saying that this sort of approach is
advocated by the likes of Tooby and Cosmides. Have I got this hopelessly
confused? I haven't had time to read much more than the introduction, but
I'd be very grateful for any clarification; does he really intend the
latter approach, either exclusively or in conjunction with the former?
One problem is that the introduction is (obviously) trying to set problems rather than solve them, which is what the rest of the book does!
There are two major problems for understanding;
firstly, that models may be set out in a black-and-white form to make their point clear, even if authors are aware that everything is shades of grey;
that the optimal solution depends on how regular the environment (including the internal environment provided by the rest of the body) is.
We may contrast, for example the human eye-blink and computer mouse-handling skills.
The eye-blink comes close to a reflex/instinct, and is 'instructed' whenever the 'challenge' of a looming object appears. Similarly, antibodies are 'selected' by the appearance of their particular antigens. What has happened is that evolutionary selection has built in lock-and-key responses - it seems a bit silly to say that development 'selects' the blink when natural selection has given it no alternatives.
For something like development of the visual receptors or speech there is more flexibility, so that radically different environments produce radically different outcomes. (Note that all humans grow up in fairly similar visual environments and end up with similar visual systems - if everybody spoke the same language we would be impressed by how inflexible speech development was - in each case the child ends up with a rigid fit to the local environment). In the case of the visual receptors, the selection must be fairly strong so the baby/kitten picks out horizontals/verticals even when there are poor examples around - otherwise only babies whose cots had vertical bars and horizontal rails and who were laid on flat mattresses would grow up normal! The evidence for speech is also that parental speech gives poor examples of accurate grammar - parents also usually accept/reject what the baby says on the basis on meaning, not grammar. So the baby seems to be selecting correct grammar from very unpromising material.
So the point here is that what looks flexible isn't - in the same way that a car will get you anywhere in the country - provided it's on the road system.
On the other hand the selection may be a bootstrapping process, so the alternatives emerge from previous stages of development and are not represented in the genes i.e. they are something like; if there is sufficient material, make a finger/whisker if the finger/whisker is stimulated. reserve a patch of brain tissue for it if two adjacent patches are stimulated, co-ordinate the responses.
So here the selection processes are much more general (rather in the way that mathematical/statistical solutions are general and can be applied to a wide range on individual problems which have the same design - the analogy also holds that in designing a psychological experiment what you want to do tends to be 'selected' to match what the statistics can analyse - reality has to be bent to fit the methods!).
So, for example, mouse-handling skills (in my experience) draw on drawing/handwriting skills - the environment has selected a variety of motor skills for dealing with objects of different physical characteristics, and on encountering a new object you match it to existing objects you know. (Note that technology is selected for being easy to use!).
Hope this helps!
REFERENCES ON CO-OPERATING GENES..
.which I dont think I specifically mentioned. Zimmer (week 1), Goodwin (week 2) and Michel & Moore (week 2) all discuss the way in which genes interact as I described - it's standard stuff which was old hat when I was a student so you'll find it in most biological textbooks.
All these make the point, which also applies to memes (see week 10 discussion) that though it's useful to think of individual genes having an effect, in fact they interact in groups. See also the next item on generative fields.
I was asked after Monday's lecture about the relation between gene control and (my memory may fail me here) the ideas John Pickering has mentioned of 'generative fields'.
In replying, I referred to week 2 and especially Goodwin's 'How the Leopard changed its spots'. Goodwin's idea is that much of development is emergent and self-organising - as I indicated for limbs, Goodwin suggests that the developmental field divides the developing bones into the appropriate number of fingers and that a gene affecting finger number only affects the amount of material the system has to deal with; less material fewer fingers.
While Dawkins wouldn't agree with this he would describe the gene as a 3-finger gene because that is an easy description of what it does.
The following correspondence between me and John Pickering ensued:
My comments on John's comments - and with grateful thanks for the
references! Many thanks, John.
Sean
Dear Sean and fellow cybernetic scholars,
Thanks for the interesting email about the idea of there being genes
specifically "for" things
As I understand Goodwin, and before him Waddington & Piaget, the role of
genes may not be to specify what appears in the genotype but rather to
permit structures to appear, given the right circumstances. That is, genes
are necessary but not sufficient precursors of organic forms.
I think this idea goes right back to D'Arcy Thompson circa 1917 - he argued that a lot, especially of lower animals (and the early stages of higher ones) simply made use of 'easy' forms derived from physical processes. Thus they tend to be circular in cross-section because this is the natural shape for a structure which relies on surface tension or hydrostatic pressure within a membrane - this doesn't need genetic instructions any more than a balloon needs instructions to be round. More complicated animals are less influenced by their materials, but phylogenetic inertia means that their evolution and development start from a point of what the materials will allow. This applies to brains, in that nerve cells and muscle cells start from a very similar design and in fact use similar chemical processes to function - the analogy being that all computers had to be mechanical because that's how Babbage started. Genes can only work within these limitations.
The conventional view is that the environment cannot directly affect
the genes - but if three fingers are an advantage, genes which reduce
the amount of precursor material will be selected for.
Mae-Wan Ho claims that there may be direct environmental influence on genes
(see: Ho & Fox, Evolutionary Processes and Metaphors. Ch. 7 'On not
holding nature still'). As I recall the evidence for this was clear, but
confined to plants and bacteria.
Must look at this - the conventional view being that the environment has no directional effect on mutation - i.e. that useful mutations are not more likely than harmful ones and that all mutations are random in regard to their selective effect.
In a rather similar way, if weaver-birds are selected for good
nests, brains which readily learn to build well will be selected; if
the pressure persists over evolutionary time, the ability to produce
a perfect form as quickly as possible in life (i.e. with minimum
wastage of time and effort) will be selected for. As the experience
needed gets less and less, the situation approaches instinct i.e. the
form of the brain is genetically fixed in the right pattern. There is
a rather mathematical paper by Maynard Smith which describes this.
Note that here, selection an ancestors in real time leads to genetic
change in evolutionary time - genetic variation in the direction on
instinct is selected for.
Back in the early part of the century the
explain how genes might track culturally learned patterns by making them
easier to assimilate (see: Russell, The acquisition of knowledge.).
Gottlieb and Johnston coined the horrid phrase "Neophenogenesis" (see:
Johnston, T. & Gottlieb, G. (1990) Neophenogenesis. Journal of Theoretical
Biology. Vol 147, pages 471 - 495.) ; Piaget called it "Phenocopy" (
see: Behaviour & Evolution. Esp. Ch.s 1,4,6 & 9) and Waddington called it
"Genetic Assimilation". In chapter 1 of "Self-Made Man", Kingdon deals
with the same idea.
Maynard Smith wrote a very short article on how connectionism provides the
theoretical vocabulary to talk about all this (When Learning Guides
Evolution. Nature, Oct. 1987, Vol. 329:761). More recently this specific
idea has resurfaced in a book called "Rethinking innateness : a
connectionist perspective on development " edited by Elman, Karmiloff-Smith
and others.
This is the Maynard-Smith reference I was thinking of. An interesting point is that one would expect selection to eliminate learning on this model - it doesn't because life-time flexibility is worth its cost in a changeable environment.
The cultural analogue is messier; whereas it is conventional wisdom
that genes cannot be directly affected by the environment during
transmission for parent to offspring, and that selection can only act
on genetic variation which arises without reference to the
environment, memes can be altered during transmission, and the
learner may understand them differently to the teacher.
Therefore while 'the gene for' is a reasonably accurate, as well as
useful, simplification of the genetic generic field-
'the meme for' is a much greater over-simplification of the cultural
generic field, though in my view a useful one for analytical
purposes.
Any comments - John?
Clearly, simple analogies between cultural and biological inheritance won't
do. I never understood why Popper had such high regard for the patently
daft views of
philosophy of Karl Popper, edited by Schlipp, P.). Dawkins is right to
broaden our ideas of what inheritance may mean, but then he's rather been
overtaken by recent revisions to Neo-Darwinism (see Edelman's "bright air,
brilliant fire")
Drastically over-simplifying a 'gene for' say long wings would be represented in wing structure (allowing for environmental effects, as discussed below) - if I understand Edelman, a 'meme for' the idea of wings would be represented in brain structure i.e. the connections made as a result of neural Darwinism. Both depend for expression on being placed in a suitable developmental environment i.e. a gene for long wings would have no effect on human development as it would not mesh with the rest of the genome, and an explanation of wings in Japanese would have no influence on English speakers because the concepts would not be intelligible and therefore would not mesh with those the English speakers possessed.
I'm not too sure about what a 'gene for' something complex would actually
do. For simpler things the idea has some content, especially when we're
able to trace the causal metabolic path (as in eye colour or
phenylketonuria for example). Once one begins to think about genes as
permissive rather than than prescriptive, then the idea that there could be a
gene 'for' anything as complex as intelligence or sexual preference feels
far less believable. Finding that a gene is more common in certain
populations merely demonstrates co-occurence or correlation. This is not
causation, as we conventionally learn in statistics classes.
Tracing the path between gene and behaviour is obviously extremely complex, especially where development of the behaviour depends on interaction with experience (as for sexual preference or language). I am persuaded e.g. by Pinker's 'Language Instinct' that children are capable of more complex constructions than their linguistic environment gives them the evidence for - though there is no realistic possibility of tracing the path from the genes, this implies to me that e.g. the creation of a creole comes from children's genetic 'knowledge' rather than the linguistic environment (a simpler pidgin).
The description of a 'gene for' e.g. sexual orientation when there is continuous variation is I think due to two things;
1) a simplification of the real situation for modelling purposes -
2) the fact that, whatever the underlying motivation, an individual who fails to make a choice gets no partner - so a weak preference can't be told from a strong preference very easily at the time of choice (it might show up in successive choices, for example in bisexuality).
AN INTELLIGENT QUESTION ABOUT INCEST
A query arising about the video clip which I use to make the point that people recognise those they are brought up with as sibs (and as models for the genetically similar people you ought to choose as partners), but others as potential partners - so that sibs brought up apart fall for each other - because they fulfil both criteria (similar but brought up separately).
If you are brought up apart from your sibs, how do you know that you look like your sibs?
Several possible answers (the tape does not give the details to answer exactly how the people involved were brought up);
Of course the video tends to concentrate on the dramatic cases of incest, so the animal work, where conditions can be controlled, is the more reliable.
In addition, the mechanism is 'cheap and dirty' (otherwise it would include a 'conscious over-ride' so that people who was aware someone was their sib could decide not to find them attractive, and real sibs are often not that similar, so the actual situation is not as dramatically predestined as the video seems to imply.
ANSWER TO QUESTIONs ON THE EVOLUTION &
DEVELOPMENT ESSAY
1)
The quote which needed explaining
To what extent is individual behaviour constrained by a genetic heritage
taken of populations? Second, to what extent is the population heritage
influenced by the behaviours of individuals?
Answer (1):-
This is a little difficult to interpret without the wider context but an
example might be helpers at the nest - as I said in the lecture it's more
helpful to compare helper and non-helper genes to see which is fitter rather
than trying to work out whether it was the parents, first-generation or
second-generation helper who was actually responsible as an individual for the
greater success of HATN. But of course (a problem I didn't raise in the
lecture) how does selection actually act if each is contributing a slightly
different role? E.g. parents normally drive the first brood away and the first
brood normally then raise their own young - so is selection actually acting on
HATN young for helping their sibs or on parents for not driving them away? And
in the
Second quote:-
We would try to account for behaviours in terms of the structure of the
nervous system, hormones and so on, as the sum development and environmental
expression of the inherited mechanism.
In turn, how proximate mechanisms have evolved as a result of selection
might explain (or, be explained by) the ultimate cause of physiology and
behaviour, in terms of the genetic history of the species population. The
question asked is to what extent are individual behaviour shaped by
evolution?
Answer:-
Obviously the questions I asked above have to be answered in terms of developmental and physiological mechanisms which makes it even more difficult, especially when species have flexible responses - it's very unlikely that there was any genetic change between the Seychelles warblers which had HATN on their original island, the first generations on the new island which did not, and the later generations which did.
A possible way of solving this problem is to think of an analogy. If you stand on a pile of bricks, which brick is holding you up? You're being supported by the whole population of bricks and the individual behaviour (fairly simple!) of every brick. But the pile was built sequentially and you could have stood on it at any time during the process though it would have been less high (= complicated) in the early stages.
So for HATN, nest-building must have evolved first (in early reptiles), followed by incubation, then care of the young (maybe in dinosaurs). So young adult birds have a propensity to care for chicks. In most species the parents chase them away before they can display this to sibs, but if the parents are tolerant they can stay to be HATN. So the critical gene for HATN is likely to be parental tolerance, though, as in the Seychelles warblers, offspring, whose parents are tolerant, may be switched to be more 'clinging' if they are underfed (based on Chisholm's point that humans become more assertive if parents are less contingent - because the parents are too busy with their own problems to attend to their children). The point I'm making is that each stage builds on what is already there - if the young adults weren't already likely to feed chicks, parent tolerance would have no selective advantage. So the critical gene for HATN is the one which allows the system to capitalise on what already exists - just as you actually stand on the top brick but wouldn't be so high if it didn't rest on the others which were already there.
I should say that my HATN explanation is speculation but the general model of identifying which are the widespread and foundation adaptations and which the later ones seems valid. For example, most primates have long-lasting personalised social relationships, continuing family bonds etc.even if they don't show evidence of having a theory of mind. So a theory of mind is likely to be an add-on which is oriented to social intelligence, and speech is likely to serve a socially-oriented theory of mind. So it's likely that the selection was between more and less effective speakers which were otherwise fairly comparable. Selection can only work on differences - if all members of a species have something in common which is disadvantageous all selection can do is eliminate the species.
The problem of which of the multiple causes of a piece of behaviour is THE
cause is a problem which has certainly been round since the 60s. Equally,
behaviour often has several functions - I think Marian Dawkins' first chapter
discusses egg-shell removal in gulls - the broken shells could attract
predators, hurt the chicks, or harbour bacteria. In fact cliff-nesting
kittiwakes don't remove the shells so only predator avoidance seems a powerful
enough selective pressure. Where natural experiments like this, or actual
experiments are not possible, ethologists have gone in for conceptual
experiments like
2)
> Having thought perhaps too much about the development and evolution
> essay, particularly focussing on the word 'connected',
You may be reading more into this than I thought of when I was (hurriedly) writing the question - what I was thinking of was that both are gradual processes (given my point that legs and language are relatively small changes with far-reaching effects)
> I guess I'm
> probably going down the wrong road with thoughts of 'adaptations and
> the status of natural selection'? i.e. - 'does behaviour have a
> special role in the evolutionary process'
As I've said on the website the point about behaviour is that it's not such a commitment as structure i.e. an individual can show a behaviour for part of the time (as in my last post about callers and satellites - a caller can decide to shut up, whereas a peacock or stickleback can't switch offers conspicuous structure). So behaviour offers the possibility of exploring a new niche in evolutionary time; if it works a firmer commitment can be made via structure to support the behaviour. E.g. the point I made in some recent comments on the big brains essay - crocodile brain size has apparently decreased, perhaps because initially more flexible behaviour (requiring brain processing) has been replaced over evolutionary time by more specialised structure which will do the same job more effectively.
(and does the evolutionary
> process have a special role in behaviour) and, 'what sorts of
> behaviour might be explained in terms of fitness'.
>
> I'm a bit thrown by the sections of Huntingford that you recommend for
> week one. We seen to have lots on 'technicalities' such as measurement
> fitness and models of adaptive significance etc (which seem to cross
> over into 'optimality'). Can say a bit more on what you might be
> thinking of for this essay?
Errm - I think there's a general point (which applies to other Qs as well) that the references tend to remain stable from year to year - the areas pointed out don't apply exactly to the particular Qs asked for a given year, which are set later on. But looking at Huntingford, Chs 6/7 especially make the point that aspects of evolutionary change are gradual and could have been achieved by gradual change in the developmental predispositions of the species concerned. Huntingford is also usefully critical (in Ch5) of 'Panglossian' attempts (to which I'm rather prone) to apply evolutionary models to everything which is why her book is still valuable, if now rather out of date.
3)
> One approach to the evo-devo essay, as it occurs to me, is to give
> first blush definitions of evo and devo i.e.organisms develop
> gradually through gene x env. interaction, so producing p'type by
> which organisms evolve gradually to produce a change in the dev.
> programme itself - hence the possibility for connection over
> evolutionary time.
Yup
We also have a further sense of connection in the
> idea that ontogeny follows phylogeny in development time - though not
> in strict terms, which I presume takes up a lot of the essay i.e genes
> and behaviour.
Yup - the reason behind this is the cranes and cathedrals I mentioned in lecture 2 - individuals have to build with the materials they already have (which in evolution will be the ontogeny of their ancestors. Hence there is a lot of 'Irish' I wouldn't start from here if I were you in development.
And of course the statement that evolution is gradual
> can be questioned along Gouldian lines.
Brown / Segestrale (core references, page 5 of your list) may be useful. Also what I've mentioned in the other email about fingers, HATN and language. This is one of the more 'angels-on-a-pin' areas of the subject.
All that remains is to write
> the essay.....
>
Not volunteering!
4)
A couple of points which came up in a discussion of an evo-devo essay;
Both evolution and development are continuous 'flow' processes i.e. cntinuing existence unchanged (of a species such as the coelacanth or on an adult individual) corresponds to 'constant speed' rather than a stationary state. Evolution / growth corresponds to acceleration, regression (the progressive reduction of the appendix or wisdom teeth, or of brain size in domesticated animals and parasites/sedentary animals which have evolved from free-living animals such as tapeworms and barnacles) corresponds to deceleration - i.e. less 'power' is being put in to oppose the selective forces of 'cost'. Thus domestic animals don't have to comprehend the world as well as wild animals because they are fed / protected and their (expensive) brains don't develop so much. Punctuated equilibrium therefore would correspond to alternate periods of acceleration and steady speed rather than evolution 'stopping'.
Evolution tends to 'add on' complexities at the end of development because, with the boot-strapping nature of development, late changes have fewer knock-on effects than early changes - but the evolution of speech shows that a major evolutionary change can occur early on. (There are several examples where inverebrate groups whose design is 'simpler seem to have evolved by changes in early development - for example centipedes start with few legs and add more in development - insects may be centipedes which never grow the extra legs (primitive insects such as silverfish have no wings)).
5)
> I understand what the question is asking, but my
> problem is where to start!! Do we have to include
> information regarding human development and evolution?
No - humans are just one species and you don't have to include them - but you can if you want to. (I don't normally ask questions which specify 'human' or 'non-human' and - a general point - for any question - you can include or exclude humans as you wish. For humans - and to a lesser extent for other large-brained species - you may need to show that you've tried to distinguish between rational explanations of behaviour and quasi-rational rules of thumb (hence, for example the time I spent on incest between separated sibs, where the sibs know they rationally should not commit it, but do so). This problem is much less serious with early development - for example it's clear from young children's limited grasp of Piagetian skills that they couldn't deduce interpersonal or language skills from their knowledge of objects - so the emergence of these skills must derive from pre-existing acquisition rules. (Note that if the environment is sufficiently predictable the genes need only specify what needs to be selected from - mature skills will emerge via neural Darwinism and input from the environment).
Basically I'm saying that you can use examples of emergent abilities from developmental psychology if you think they are appropriate.
Note that I said there could be apparent developmental/evolutionary discontinuities e.g. finger number, helpers at the nest where the apparent differences between states are great but accounted for by relatively simple developmental switches. I'll mention another example in week 9 - Worden (chapter 9 of Hurford et al 'Approaches to the Evolution of Language') suggests that most of the evolution of language is accounted for by Machievellian intelligence. He suggests that in order to predict the behaviour of others effectively it's necessary to have implicit labels for your own behaviour and the reactions of others e.g. a monkey can predict that if it bites monkey X it will be bitten itself, that if it grooms it will be groomed. In evolution (and in human development) individuals understand these general rules first, then label them with words - but words increase the rate of cultural evolution by orders of magnitude.
> As I notice that some of the references for the
> reading point in that direction.
>
> I hope some one out there might be able to give me a
> few ideas. I hope my question doesn't seem too
> obvious, as maybe I have missed the point.
No!
Finally,
> when this email reaches everyone, I apologise for the
> name it says it has been sent from, someone else set
> up my account.
Should be possible to change it - can anyone advise?
6)
> regarding the evolution and development essay, i am having difficulty
> understanding what the question is asking. ("both developmental and
> evolutionary change are gradual. Are these two facts connected?) To me
> it appears that the emphasis of the question is if the fact that the
> two are gradual and if this links the two. If this is the case i am
> having a lot of difficulty finding specific reading on this. Also from
> the other comments posted about this essay it seems as though the
> gradual bit is less important and that it is just factors possibly
> linking development and evolution which are important. Please clear
> this up for me as i really don't know how to tackle the question. I
> have some ideas of what to say but am having trouble finding specific
> references to back them up
>
This is mainly a book-based essay, though you might find some journal articles in, for example, Behavioral & Brain Sciences. The references are mainly in weeks 1 & 2;
Gazzaniga and Edelman (week 2 - Gazzaniga also week 1) give the clearest links between gradual evolutionary change and gradual brain develeopment, Michel & Moore (week 2) are also good in a more textbooky way.
Zimmer and Dawkins 'Climbing Mount Improbable' (week 1) give examples based more on structure but with more behavioural implications (e.g. Dawkins on spider webs); Goodwin (week 2) deals especially with emergent organisation in development.
Many of the other references in weeks 1 and 2 touch on gradual change, as does, for example, chapter 6 of Majerus (week 6) - the Red Queen, for example, depends on gradual change, especially in the structure chosen (e.g. the development of a progressively longer tail).
You'll notice quite a number of these have a structural basis as behaviour doesn't fossilise - though reverse engineering has been used to predict behaviour in, for example, 'Walking with Dinosaurs' (I notice there's now a book out about this).
To some extent it has to be an act of faith, given that both behaviour and embryos fossilise poorly, that gradual change in current behavioural embryological development relates to gradual change in evolution - though young dinosaurs etc. have been found which show the same baby-to-adult changes (chubbier babies etc.) which we see in modern vertebrates.
7)
> I KEEP GETTING CONFUSED AS TO WHAT THE QUESTION ON EVOLUTION AND
> DEVELOPMENT IS ASKING EXACTLY. IS IT STATING THAT DEVELOPMENTAL AND
> EVOLUTIONARY CHANGE ARE GRADUAL AND THEN ASKING WHETHER EVOLUTION AND
> DEVELOPMENT ARE LINKED?;
This would be the most likely approach.
OR IS IT ASKING US TO DISPUTE THE FACT THAT THEY
> ARE BOTH GRADUAL AND THAT INSTEAD EITHER ONE, NEITHER OR BOTH MAY ACTUALLY
> INVOLVE PUNCTUATED EQUILIBRIUM OR SOME OTHER ALTERNATIVE?
This would be a valid alternative; as I said in week 1/2 and also in week 10, relatively slight changes in development (legs and language) may have extremely far-reaching effects so gradual change in ontogeny (development) may lead to major change in phylogeny (evolution).
>
> PLEASE COULD YOU GIVE ME SOME IDEA ABOUT WHICH ASPECT TO CONCENTRATE ON AS
> I DON'T WANT TO GO COMPLETELY OFF THE RAILS!
The essay questions I set are intended to allow a range of acceptable answers - there is no single 'right answer'. See also my other posting on this question which I have sent to the ng. Provided you have got the basic concepts correct (as you seem to have) and can produce a coherent argument
you should be OK.
NOTES ON ESSAY DRAFTS BOTH DEVELOPMENTAL AND EVOLUTIONARY CHANGE ARE
GRADUAL. ARE THESE TWO FACTS CONNECTED?
1)
*If you are limited for words you could cut out this section as your reference to Ridley covers these facts. A critical aspect is that major mutations are more likely to disrupt development and kill the animal; slight alterations will be viable and can therefore be exposed to natural selection.
*Good point - I assume this is the peppered moth - the dark morph was already present before industrial pollution and its increase and decrease shows how the change in gene frequencies tracks selective forces.
*The current concern about global warming (and other human effects) are that they are too rapid, in evolutionary time, for species to respond to.
*This and first part of next sentence assume purpose in selection in fact most species in time meet a selection pressure they cannot deal with and go extinct e.g. the ichthyosaurs seem to have been outcompeted by the plesiosaurs.
*Stress the 'usually' -as you say below.
*later
*Good
*One cell - the fertilised egg.
*Good - though Piaget's detailed work has been challenged by later research the general point that younger children cannot perform certain tasks remains true.
*Proof-read here and in next sentence
*Because of this nature of development, evolutionary changes can more safely be made at the later stages of development; mutations at early stages are likely to disrupt all later stages, except where the nature of development allows alternative pathways (e.g. the later development of fingers is not disrupted by an early change which changes the number of fingers - Goodwin / Zimmer).
*This is Dennett's distinction between 'Darwinian' and 'Skinnerian' animals; the former adapt by the death of individuals who are inappropriately 'hard-wired' for the prevailing conditions, the latter by individuals changing to meet the requirements of their habitat. Flexibility has a cost (in big brains, parental care etc.) which is only worth paying by species which have fairly good individual survival prospects - e.g. most insects are so vulnerable that they cannot afford to 'waste' time learning (bees are an exception probably BECAUSE their learning benefits the invulnerable colony).
*This section repeats points made earlier
*Still repetitive.
*Rather repetitive - if a particular developmental change is repeatedly selected for, mutations which make it easier will be favoured so genetic predispositions favour a particular type of environmental responsiveness - i.e. what was originally learned with difficulty will become learned easily or even hard-wired, despite the resulting costs in flexibility e.g. the example I gave of the primate ease in learning to avoid snakes (the visual cliff is another example, where species such as humans which are at risk of falling avoid the cliff without any experience, species such as lambs do not)
*Hormones probably belong earlier - they are a very 'primitive' form of communication (plants have hormones but no nervous systems) and can be produced in variable quantity, so response can be 'tuned' readily - also response of target cells can be 'tuned'.
*You need to cut out repetitive sections, which will give you the elbow-room to add introduction and conclusion. While physical development is necessarily gradual (simply because every cell in the body must come from divisions of the one original egg-cell) behavioural development does not have to be, but as you point out, individuals gain from flexibility if they can bear the cost of the 'reserve capacity' for flexibility.
2)
*'scare quotes'
*Reference?
*Not clear what last sentence means - otherwise good.
*Reference?
*Good
*Should this be Zimmer?
*Gould and Goodwin are both critics of the Dawkinsians though from rather different viewpoints Gould is criticising evolution as 'steady process' - Goodwin is saying that the influence of genes is over-rated as physical processes have considerable effect on development.
*Very good point - reference?
*Phrasing in last two sentences needs tidying up.
*Does Ridley actually mean that people (say) can choose their own genes or (e.g.) that by contraception people can select against parental-behaviour genes? (Note that if you have Ridley's quote correctly, it is objectively true i.e. you can't be criticised for quoting Ridley correctly.
*Good -reference
*Phrasing needs clarification
*Is Lorenz a secondary reference?
*Do you mean 'in early life'?
*Interesting points here but you need to clarify
*Reference for Dennett?
*orchards! And I don't think they ate a whole orchard ; )
*Reference?
*Not clear what this means and reference?
*Ditto
*Yes - reference?
*Good - reference?
*Good -reference?
*Good
*Reference?
*Clarify - reference for Conway Morris?
*Reference - there are two books for 'Walking with Dinosaurs' - the programmes suggest it was simple hunger-related cannibalism rather than pride takeover.
*Good
*Conclusion needs tidying up.
*Overall, argument needs tidying up as marked.