animal cognition Flashcards
what is cognition (7)
basis for intelligent behaviour
core vs higher-order abilities
effort and attention required - no autopilot
top-down mental processes
controls sensory, memory, and motor systems
PFC in mammals
overrides reflexive, habitual response in favour of complex, long-term goals
core/basic cognitive abilities (3)
working memory
inhibitory control
cognitive flexibility
higher order/insight related cognitive abilities (5)
object permanence
self-recognition
mental time travel
ToM
tool use/causal reasoning
why study animal cognition
see if they have some abilities same as humans
then connect to neural mechanisms
scala naturae - what is it
aristotles ladder of being - hierarchical structure
degree of perfection – compare where we stand compared to other species to understand our cognitive ability
find the point at which animals stop being instinctual and become introspective
scala naturae order (5)
mammals
birds
reptiles
amphibians
fish
birds cognitive abilities
working memory - recognising images
very intelligent species
reptiles cognitive abilities
inhibitory control - stop response to gain things later on
amphibians cognitive abilities
cognitive flexibility - adapt to new rules
learned inhibition - stop responding to certain stimuli
fish cognitive abilities
working memory in zebra fish
- correctly select a previously shown colour in order to receive a food reward
what we learnt from animal studies of cognition
insight related cognition may not just be in mammals
underestimate other species as they are viewed as lower down the hierarchy
evolutionary distant birds and mammals share these abilities - must have similarities somewhere
lamination of mammal cortex
6 layers
differentiation of neocortex allows humans to perform cognitive skills
prefrontal cortex
association with core cognition and executive function
damage = deficits in executive function
can dissociate between EFs using localisation of lesions in patient studies
bird vs human brain
birds have no neocortex – supposedly cannot perform cognitions without it, so maybe it isn’t that special
humans are introspective on their own cognitions
define working memory
The representation of items held in consciousness during experiences or after retrieval of memories. Short-lasting and associated with active rehearsal or manipulation of information (Miller, 2000)
mammalian PFC for WM
how to study animal brains with electrodes
electrodes measure action potentials and depolarisation
invasive - in brain not just on surface
- electrodes measure voltage changes in extracellular space around neurons
- allows to measure frequency of spikes as action potentials
- visualise with spike-trains - dots and dashes
- precise timing of action potentials - spatial accuracy
delay response task in monkeys - Niki (1974)
method
* cue light illuminated (left or right), then 2-3 sec delay, then monkey has to choose left or right
results
directionally selective neurons:
* become active depending on which side the light was on
some are directionally selective during delay period
* activity spikes after cue and drops after choice was made
* sustained delay activity
* info is encoded until it is needed and then is abolished
some drop in activity during delay:
* spike at cue, then drop in delay, and increase again following behavioural response as choice is made
* inverse of delay selective neurons
delay response in monkeys - what is learnt from this
delay activity = neural basis for working memory
need to research what enables this delay activity
dopamine patterns and classical conditioning
before learning = DA increase with reward
after learning = DA increase with conditioned stimulus and at baseline with reward
with reward omission = DA increase with conditioned stimulus and then DA decreases when reward is meant to be
dopamine as a neuromodulator
responds to sensory stimuli which predicts future rewards
signal can be used to tag sensory cues as relevant and facilitate their entrance into working memory
DA and delay response task
cue left and right leads to response in DA neuron
this increases DA in PFC
this enables persistent delay activity in PFC
does delay activity show working memory
not quite - WM is not passive - info must be manipulated
therefore the cue light study in monkeys could be measuring STM not WM
cortex differences in human and birds
birds don’t have a neocortex - used for cognition in mammals
have subcortical structures - basal ganglia, globus pallidus etc
previously thought to be comprised of structures similar to basal ganglia - primitive structures
smaller areas said to be like human cortex
BUT change in opinions
larger areas now thought to be more similar to a human cortex → pallial area
from perspective of structural similarity
neural similarity - what regions do is more important in understanding brain differences
NCL in birds
nidopallium caudolaterale (NCL)
- pallium = term for grey and white matter covering cerebrum
- nidopallium = nested pallium
- previously thought of differently as more primitive
- nidopallium has subregions - e.g. caudolateral
avian NCL and mammals PFC similarities
analogue - same/similar function even if structure is different
directed forgetting study in homing pigeons
Rose and Colombo (2005)
A = pigeon has to remember
ITI = inter trial interval
sample = stimuli presented - animal pecks cue
remember cue = audio indicates they need to remember the cue
3 second delay
test period = select sample from 2 choices - comparison
provided reward on correct selection
similar to STM tasks - more elaborate so choice is required here
B = pigeon has to forget
given auditory cue to forget the sample
need to discern whether to remember or forget
results:
A - remember
sustained activity
increase from sample - consistent from cue to end of delay
as the info is needed to make a choice in the task later on
robust across other findings
B = forget
delay activity is abolished after the cue
activity drops off after cue once they know they don’t need it anymore
evidence that NCL has delay activity → participating in working memory
limitations of homing pigeons delay activity study (3)
- careful with interpretation
- reward prediction, not delay activity
- abolished delay activity could be linked with reward prediction - food reward given
- no activity as they realise they wont get a reward for it
- discern between WM and reward prediction → same neurons could be involved in both of these - not purely discrete from each other
- motor prep to get food, not WM
- activity seen could just be preparation of motor control in response to task to get food reward - not necessarily working memory
working memory in crows study methods
delayed match-to-sample task (MTS)
method:
go stimuli
pre-sample delay
sample presented (image) in centre of screen
delay
choice made between images on different corners of screen
allows insight into WM:
* disentangle WM from motor preparation - location of stimulus changes randomly
* reward not always given → chance of reward was equal for all match items and randomised → therefore reward expectation negated
crows are very good at this task - cognitive task
working memory in crows - sample-selective neurons
each neuron represents one of the 4 images presented
different levels of activation for different stimulus → distinguished 4 sample stimuli with firing rate
therefore are manipulating info → different processing of each → this is working memory
response of neuron = active during selection period
selective discrimination of stimuli - non-arbitrary
working memory in crows - delay-selective neurons
response of neuron is selective during delay period
not distinguished in sample period
lower activity in S4 here
more than half of the recorded neurons are either sample or delay selective or both (shown above)
relation between neural activity and task performance
perform task very well overall
error trials → firing rate does not distinguish the different stimuli as well as in correct trials
true for both sample and delay selective neurons
evaluation of methods for studying working memory in crows
advantages
allow direct recordings of action potentials – single cell recordings
not usually possible in humans - invasive
spatial and temporal accuracy can’t get from EEG, fMRI
insights into the nature of cognition, analogues of cognition in different species – basis of intelligence
limitations
invasive – can be stressful for animals
animals cannot self-report, have to be trained on tasks - operant conditioning
therefore tasks do not reflect ‘typical’ behaviour → do they actually have abilities or have they been trained to respond to tasks
difficult to design tasks to remove confounds – reward anticipation, motor preparation, STM vs WM?
bee brains
only 1 million neurons
large compared to fruit flies
not simple, hard-wired structure
honey-bee physiology, navigation, and communication
working memory in bees study
Zhang et al (2005)* - visual working memory in decision making
delayed match-to-sample task
method:
* bees fly into a tunnel
* part way through they see a target sample (square pattern)
* then fly to the end of the tunnel where square patterns are presented where they need to choose the target response
* correct response = reward
* distance between sample and choice can be manipulated to understand working memory
* varied time interval = varied difficulty
results:
short delay (1.24s) = high performance (75% correct)
performance showed working memory of up to 6.5secs
transfer studies of working memory in bees
presenting incorrect patterns in the tunnel
either before or after sample stimulus
transfer means using the rule e.g. “correct pattern first” when encountering novel stimuli
correct pattern is always presented at same distance from entrance
baseline = without incorrect patterns
transfer 1 = present incorrect before correct
transfer 2 = present incorrect after correct
- do well in all these
transfer 3 = correct pattern moved to IP1 and IP2 position
- do worse in this
inferences from bees working memory
potentially shows working memory - could be short term memory → no info manipulation needed, so is it just passive STM
bees have cognitive abilities despite small brain size
bees may have working memory of 6-7s → unclear to what degree active memory content can be manipulated
cognitive flexibility experimental measures (3)
ability to perform a large range of behaviours and tasks
select flexibly to suit different contexts → adapt to new rules to achieve goals
experimentally:
* questionnaires
* response to sensory stimuli
* following and adapting to task rules
crows cognitive flexibility studies
delayed match-to-sample task (Veit et al 2014)
present stimulus → delay → cue (visual or auditory - determines what is needed to be done with info) → select correct choice
trial type 1 = match-to-sample = pick stimulus that was shown
trial type 2 = non-match-to-sample = pick stimulus that was NOT shown
assume this requires core cognitive functions → PFC in humans and therefore NCL is used in birds (analogous)
crows performance is high on this task
rule learning in crows
high firing rate in “non-match” rule trials
low firing rate in “match” rules
doesn’t matter if cue was auditory or visual
shows response to abstract rules on single cell level in NCL
supports cognitive flexibility in birds
supports NCL as analogue to PFC → is cortex actually special then
cognitive flexibility in bees
experiment as above with bees flying through a tunnel
shown 2 patterns → some the first is the correct and others the second
both groups successfully learn which is correct and select correctly at the end of the tunnel
learning tests show bees learn the rule and perform well on trained stimuli
transfer tests introduce novel patterns (not known to bees) → still perform above chance levels, but not as well as in learned task
can transfer the cognitive learnt stimulus to learning the rule instead
indicates bees generalise from specific visual stimuli to more abstract tasks
learning of abstract concepts in bees - training
bee has to navigate through a “maze” to the feeder
they are presented with an odour → lemon or mango
they must use this odour to navigate to the feeder - match to it
e.g. if they follow the lemon scent they’ll get to the feeder, mango would lead to nothing
learning of abstract concepts in bees - transfer test
after training - odours replaced with colour
test if presented at first with a colour, will they know to generalise the odour rule to this and select the correct colour out of two presented later on
results
perform above chance level in delayed-match-to-sample task
can generalise from odours to visual stimuli
above chance level in cross-modal learning
shows abstract concept of “sameness”
further experiments by these people show bees understand “difference” too
object permanence study in crows - 2 trial types
object permanence as a high order cognitive ability
Hoffman et al (2011)
trial 1:
present a meal worm to a crow then place a cloth over it
the crow knows to move the cloth to access the worm
trial 2:
180 degree displacement
worm placed behind one of two screens and then their positions are switched
therefore the worm that was put on the right is now on the left
the crow successfully go to the left screen to find the worm
object permanence in macaques - method
single unit recordings in inferotemporal cortex of macaques
evolutionary similar to humans - less developed version of our brains
complete an object permanence task
match response - same object
object is shown, then a screen moves to cover it, then it emerges the other side
this is expected - same object should be there
surprise response - different object present
wall moves and the object would be something else
if it doesn’t understand object permanence - there would be no difference in neuronal firing
object permanence in macaques results
selective firing for unexpected object emergence
some neurons respond higher to expected, others to surprise
suggests macaques have object permanence
neural representation of object permanence processing
remember this is intuited from the research - just an indication of this
MSR in mammals (2)
mirror self recognition
dolphins
longitudinal study of one dolphin
approaches mirror in same way as a human child
engage in behaviours (very physical) as they move and attempt to understand how their own movement looks and what their body can do
also show other dolphins what they can do in the mirror
apes
baby initially fearful of what they see in mirror
then become curious - learn more about self and own bodily movements
MSR in birds
magpies - demonstrate self recognition → without cortex
put a coloured dot on the magpie which it will see in the mirror
then it directs action towards itself to remove the spot → if they didn’t recognise themselves they would try and interact directly with the mirror
mental time travel - scrub jays study
behaviour where they find and store foods for later → cache food
know that worms will go bad quickly so store them one way
also like peanuts - know they don’t go bad quickly so store another way
study
provide birds with meal worms and an ice cube tray full of sand
one half of tray is exposed (left) and other is covered (right)
they take the worm and stash it in the tray buried in the sand → on left side to come back for it later
then 5 days later covered the other side of the tray and gave them a peanut → cache these in the right hand side
tray is presented (but with worms removed) 120hrs later (time frame for worm to go bad)
results
bird will go back to the left side of the tray to look for the worm instantly → knowledge it would go bad soon
demonstrates episodic-like memory
remembering that they stashed the worms and where they did it and how long ago they did it
theory of mind evidence in chimpanzees
show patience with others who are willing (but unable) to share food over those who are unwilling to share
discriminate between accidental and intentional actions → understanding of intent and therefore maybe understand others thoughts and feelings
follow/track gaze of others
gestural communication when facing another - will move to face if other is orientated away
ToM task for chimps
engaging story for chimps → film a drama with a person dressed as a chimp
chimp will watch and engage with the story being shown - setting in the same enclosure they live in:
* man leaves building and sees where the chimp is
* chimp hits the man
* chimp hides behind one of two bales of hay
* man leaves and returns with a stick and tries to hit the bale with the chimp behind it
false belief condition = chimp moves and so man hits the wrong bale
eye tracking - see where chimps think the man would look
results
* if they look where the man would look - this shows ToM
* chimps perform well
* look toward target over distractor by significant margin
* demonstrate ToM - comparable with human studies
tool use in animals - what is shows
causal reasoning
tool use requires manufacturing element - create to fit a purpose
cross cultural evidence and across species (insects to primates)
use in foraging behaviours - extend reach, digging
dolphin tool use
sponges used as protection during foraging on the end of their faces
cultural transmission of tool use behaviour - show other dolphins how to do this
is this tool use? no manufacturing
crow tool use
fish for long-horn beetles
manufacture a hook-tool from a stick
tool use definition
tool use = employment of an unattached environmental object to alter another object or the user when they carry the tool. the user is responsible for the proper and effective orientation of the tool
tool kits/sets in chimps in gabon
use tool kits to access honey from underground hives → termites
modify tools - brush tools are better for harvesting insects
sequential tool use and tool specificity
limited tool use evidence
most observable behaviours foraging (by extending reach) → no interaction between tool and other objects
not specifically manufactured for purpose
tools often serve only one purpose
not intrinsic part of life for species or defining characteristic of species that engage in tool use - only specific populations do use these tools
can still use animal tool behaviour to teach us about our own behaviour
tool use in humans vs rhesus monkeys - fMRI
Peeters et al., (2009)
fMRI for observed actions and interactions with tool use to extend reach
similar activation for hand object interaction
unique inferior parietal lobe activation in humans for tool object interactions - left lateralised
anterior supramarginal gyrus (aSMG) - an area humans have that rhesus monkeys do not
different streams for perception and action
Goodale and Milner (1992)
two streams for perception and action
ventral = what
dorso-dorsal = where
ventro-dorsal = how (added by Rizzolatti and Matelli, 2003)
macaques don’t have the how stream
one species of macaque has just entered the stone age - have hammers and anvils
