Behavioural ecology > week 5 - learning behaviour > Flashcards
week 5 - learning behaviour Flashcards
Much of the early research on animals learning was acknowledged in the relatively simple form of _____
imprinting
What is imprinting?
irreversible learning that occurs in a critical time window
define learning
the process of acquiring information.
learning is durable and usually adaptive change in an individuals behaviour traceable to a specific experience in that individuals life
in learning it needs to be defined that the change is not a result of ___ (3)
developmnet
exhaustion
satiation
when is learning advantageous?
changing environments
what is stephens model for the evolution of learning?
see slide 5 lecture 1 week 5
what are the 2 forms of non associative learning?
habituation
sensitisation
explain habituation
strength of reaction decreases
with repeated, but “non-meaningful”
stimulus. Example – whistling in
prairie dogs
Explain sensitisation
increased response to a stimulus after another or the same stimulus, e.g. in Aplysia
what is an associative learning?
classical conditioning
what is classical conditioning?
conditioning of a response (reflex)
what are the 2 different stimuli in pavlovian conditioning?
UCS = unconditioned stimulus CS = conditioned stimulus
animals show an innate preparedness to learn certain things better than others, such as?
colour learning in honeybees
violet and blue flowers have the most nectar, and there is an innate preference and preparedness for learning things that are relevant.
In classical conditioning what is critical?
timing
maximum effect if US is directly preceded by CS
What is a second type of associative learning?
operant conditioning
what is operant conditioning?
learning by trial and error
animal learns to associate a voluntary action with the consequences that follow from performing it
what is motor learning?
improvement of speed and accuracy of movements
What is true cognition?
concept learning - e.g. concept of sameness/difference in honeybees
what is social learning?
learning from other individuals
example of social learning?
honey bee dance
one of the contexts under which animals learn is spatial behaviour, give the most famous example
atlantic salmon return to their native river - they memorise olfactory cues of the stream in which they were born
Until recently studies on spatial behaviour were extremely cumbersome why?
because it was hard to track animals over long distances
what can you use to track animals over long distances?
radio tracking
What are 2 of the spatial behaviour challenges required by a bee?
navigating back home
reliably retrieve locations from their memory for foraging - often multiple as single flowers do not contain enough nectar
what is the travelling salesman problem?
quickest and most efficient route around an area
when are bees faced with the travelling salesmen problem?
when navigating back home and also trying to find new locations for food
In an experiment done where bees were faced with a travelling salesman problem whereby they had to locate and remember 5 locations of nectar from the nest what were the results?
radiotracked the bees to see if they gradually optimise their solution.
here is a naïve bee. The pink arrows are the prevailing wind direction. The task for this bee is to find all these 5 flowers and link them in a way that is efficient and minimise travelling time and energy.
it goes from the nest straight past one flower to 5, then skips the nearest neighbour and goes to 2. so it is just exploring at the moment and figuring out where all the flowers are. this will take some time to make the optimal solution.
thats a number of different tracks from the same individual. track 2 is still fairly chaotic but it does find another flower. The first several bouts are random exploration. If ou look at the bee after 24 bouts it is more experienced.
this is of a bee on 28th foraging trip, it is an experienced forager and has found an optimal path. From an extremely tortuous path on the first bout it optimises its path and solves the travelling salesman problem.
top left is naive bee all others are tracks by experienced bees. Interestingly they never stick to one solution. So what they do is essentially interface exploration with exploitation. Even experienced animals often when they should know where these feeders are, interface a bit of exploration to scout around for future suitable resources and incorporate these into their path if something is found.
What is the hippocampus?
an area of the brain that is important in encoding spatial information. Of particular importance in rodents
what is the hippocampus functino in rodents?
spatial orientation
The nobel prize was awarded to researchers doing work on the hippocampus of the rats. what did they find?
explored a neuro equivalent of a physical map in animals brains.
cognitive map is the mental equivalent of what you have spatial relationships in the real world, a mental equivalent of a map. It tells you where you are in relation to land marks, moreover like a real map it allows you to travel novel shortcuts on routes you haven’t previously travelled and so on.
1970’s discovered in rodents were so called “place cells”. Cells in the rodent hippocampus that fired reliably and only if that rat was in a particular position in its known environment.
So what we have here we have panels and the activity pattern of a single cell whilst the rodent is moving in its environment which has been manipulated into an arena. Where you see blue the cell is inactive and no action potentials are produced. Yellow and red are increasing firing activity.
The top left panel shows that while the rat is moving around in a square arena the cell is only active when the rat is in the lower left corner of the arena. Now these various other panels, are the same rat in the exact same arena but varied slightly in its shape. It has been expanded into a rectangle no longer a square. No matter what the shape of the arena it only ever specifies lower left corner, that is only when the cells are fired. Not all cells are that simple some have fairly complex properties.
In the lower left group of panels. One rat, one cell, but a different firing pattern. In the original square arena that cells fires when the rat is in the middle of the top edge. That still stays the same when you expand the arena horizontally. But if you expand the cell vertically then the cell fires in a different way, it appears to encode a position relative to that corner.
Here on the other hand is a cell which appears to be closer to specifying the middle between two corners. It moves over somewhat in this particular panel.
Some cells code specifically for corners of this environment others for relationships with edges and so on. There are hundreds of cells all with individual coding. There ensemble code together encodes where the rat is in a familiar space.
in addition to these findings there are also a variety of other types of cells in the hippocampus. For example head direction cells, code for directions rather than places, as in north, south and so on. Provide an overall framework for spatial coordinates.
Intention cells in the prefrontal cortex are interesting by recording their physiological activity researchers can see what the rat will do next before it does it. These are cells that fire before the rat moves in a particular direction but will reliably predict that direction. So that the researchers just by monitoring that activity will predict what the rat will do.
the hippocampus shows a variety of specialisations that are inline with animals natural behaviours - what did CLAYTON show
caching ability in birds
background on food storing in birds
food storing in birds sometimes have extreme spatial capacity for storing information about where they are storing seeds.
these birds store many seeds a day and over a period of month has several thousand caches. They retrieve them often several weeks after caching them. It is in the benefit of trees that they forget some of the seeds. In addition to memorising where they store seeds they remember when they stored them, this is important with how different foods decay at different rates.
what question did researchers ask about birds and caching?
Given that the hippocampus has a very central function in spatial memory, the question these researchers asked then is of course is there a correlation with hippocampus and birds lifestyle as many birds cache and others do not.
What did results from Krebs et al 1989 and Sherry et al 1989 show?
There are a variety of bird families that were sampled. Relative hippocampus size is divided by the rest of the brain as want to account for variance in body size whereby larger birds will have larger brains. Black points are the ones that cache, the open squares are the ones that are non storers. IN both of these studies you can clearly see that caching species have larger relative hippocampus sizes compared to non storers. So clearly that lifestyle of storing vs non storing lifestyle has an affect on the size of the hippocampus.
interestingly it is not just a between species difference but also an individual difference, arising depending on whether you have experience with food storing. CLAYTON evidence for this?
Compared in 3 species that are all food storers the relative hippocampal volume depending on the experience of the birds the hippocampal volume. Red is before experienced, 2 groups where one was given experienced with food storing and another group of birds which were prevented from food storing. It turns out that
over experience over time has a profound influence on hippocampal volume, in the coal tit there is no change between no experience and pre experience whereas in the experienced bird there is a 1/3 increase in hippocampal size. If the birds have no possibility to store food therefore has no possibility to store spatial memories the hippocampus stays small, only for those birds that are given food to store and so must know where they did it grow their hippocampus.
Clayton also backed up his evidence on brain growth and importance of the hippocampus in food storing birds how?
to see whether it is just the hippocampus that grows or the whole brain. So what she did is to look at another area of the brain to show that it is not the whole brain that increases in size only the hippocampal region which is associated with memory.
Give an example of where an animal learns from its mating partners?
how birds can learn from observation by success in mate choices.
female bird experienced with male bird 1 (phenotype 1) and that hasn’t been particularly successful and only produced 1 egg. Encounters another male bird of same species looking slightly different, colour pattern on head, with that guy she produced 4 eggs and so remembers how the helpful male looked compared to non helpful males. She will in the future chose males based on the particular success shes had in the past and focus on having mates looking at phenotype number 2.
many mammal groups especially where individuals of limited relatedness gather together and stay in stable groups need to learn how to __ __ __
recognise each other
how can animals remember who belongs to their group?
scent markings
individual recognition of faces
what strategy of identifying members of your own group is widespread in the animal kingdom?
individuals. The very bare bones level of society all you need to remember is some cues which identify all members of the society and which don’t identify others that are not part of the society.
example of group recognition from a scent?
honey bees have guard bees at the entrance of their nest which identify every animal that enters the hive by smelling them to identify whether its scent is from this hive or an intruder that would have a different scent. In this case the identification is not by individual recognition its learning who is part of your group by common features, and this case a scent.
Example and evidence by CHITTKA of individual recognition of family members in isnects
Three portraits of particular wasps in the bottom left. These are 3 wasps from the same colony and their facial features are distinguishable, the three wasps from the same colony and same species. They have highly distinct visual facial markings. It turns out that these wasps have linear hierarchy in their colony. There is an alpha female at the top who gets to immobilise all the egg laying, if she is to die then the next individual moves up. That linear hierachy is supported by individual fights, duals which can be costly, lots of energy or sometimes death of an individual. Establish hierarchy once the individual duals and after that don’t fight as they know where they belong in the hierarchy. They avoid these costly fights. The reasons the wasps can identify each other by these individual facial markings. It is an adaptive benefit they help avoid future confrontations where an individual would definitely lose and is costly. This has been demonstrated by Chittka. Who explored individual recognition.
associated some faces with mild electric shocks and the wasps turn out to rapidly learn such facial features and retain them in memory for some period of time. They go to the ‘friend’ wasp not the ‘foe’ wasp which gave them an electric shock. Good at learning these faces and retaining them in memory. Colonies are small so the memory capacity does not have to be that large, but of course this is a distincition in memory capacity where by we have to remember thousands of faces and them only a dozen. But remember that this is a difference in capacity not quality of that memory.
Learning is also important in vertebrate aggression, fighting fish spend a good deal of time in contest with other males over territories and access to mates.
Evidence showing that the cues that identify, which predict aggression are fairly arbitrary?
Researchers made the fish associated a light signal with the appearance of another fish that could potentially be an aggressive threat. That is the green curve. As a result of that over a period of multiple weeks the fish became more aggressive just as a result of being expose to that light without actually being exposed to aggression. The red line shows the opposite. The light source is a condition stimulus which predicts the risk of an aggressive encounter.
what do you often find in species with male-male contests with regards to winners and losers?
males will remember the outcome of contests. Which in turn influences the outcome of future contests. Individuals which have one confrontation and win carry a lot of confidence and will win again. In this case it was shown in fighting fish the probability of a fish that has won its first envcounter, whether it will win its second encounter; ‘win win’ which is pretty much 100% and then lose to win will often be 0%. SO fish has lost past contest will not win the next one either. They were careful in matching the fish with same size competitors etc. So memory seems to play a strong role in the fishes chance of winning the next outcome.
This is the proportion of losses and whether you will lose again. Shows the exact same relationship as before.
What is the adaptive value of learning?
depends what environment these animals operate in
how did CHITTKA test the adaptive value of variation in learning in bees?
First tested bees in lab conditions where they learned that Blue is non reward and yellow reward.Then took these same bees and tested their natural foraging success in the field. What we found is there are profound reliable differences between learning speed in different individuals and different colonies. What is plotted here is 2 learning curves. Plotting number of errors as a function of experience.
as you move from left to right it is increasing experience, and as a function of that percent errors. This is a slow learning bee which starts at 50% errors which is normal as has no experience, takes 80 visits to have a low percent error. Fast learner starts at 50% but after 20 visits is at a low percent error. Very big differences in learning speed between these 2 animals. Can quantify each individuals learning speed by fitting a learning curve to it. Using a specific formula. T is essentially indicating how fast individuals learn, perhaps quietly counter intuitively when T is small. And not learning as fast when T is relatively low???
What did raine and chittka look at in terms of adaptive value of variation in learning in bees?
now looking at T values for various colonies. Each dot is one colony. There are some variation within each colony. Learning speed as symbolised by T value on the x axis, small T values mean rapid learning, large T values mean slow learning.
there is huge variation between colonies, some learn extremely rapidly others slowly. This is under natural conditions, testing their foraging success by measuring how much nectar they bring home. Turns out there is a good correlation between how fast they learn in lab conditions nad how good they are at foraging in natures economy.
colonies with fast learners had the highest foraging success. The slowest learning colony had the worst foraging success.
Learning in tightly controlled lab conditions is a good indicator at how good they will be at foraging in the wild.
