neuroscience of visuo-motor control

automatic visuo-motor beh

• signals that do not seem to be under control of the organism

• eg phototaxis

• we like to think that we are not influenced by this

what is phototaxis

• (e.g., moth) - attracted/move towards light

evo of the visual eyes stage 1

• Eyes did not start out for vision, simple eyes are just sensory-motor receptors, receptors in the skin/cell wall of single cell organisms that respond to light (eg how skin cells are in sunlight then respond by tanning)

• Whatever chemical change is triggered by light is passed onto a muscle cell or cilia, makes it twitch which moves it closer to the light, movement response to light 

stage 2

• Now also have cerebral photoreceptors - wired up in a way where they inhibit each other 

• So whichever receptor gets the stronger signal can shut down the opposite one 

• Means we have a very primitive form of photovision in place

• Contrast enhancement - Two stimuli slightly different in intensity will produce a response, weak difference in the actual stimulus is represented in the organism as a stronger difference

stage 3

• Then have same internal arrangement, have receptors and some cup shaped skin cells embedded, on top of that is a clear lens - Much better at capturing the photons from the sun 

• Improved photon collection allows us to become even more sensitive to the visual environment

stage 4

• Add motor neurons, sensory neurons now connected to motor neurons as well as muscle.

• Allows a switch in the type of movement that happens in response to light

• Depending on which motor neurons get more activity, the part of the body that is activated can move towards the light or away from the light (called pos or neg phototaxis)

stage 5

• Sea stars have this type of vision 

• Gets more complex by adding more neurons, eg add many interneurons 

• All of this step wise improves the complexity of visual computation and processing - eventually allows to integrate visual signals with signals from other modalities (animal might be able to smell or hear) 

• Now there are so many neurons involved, they can come together and all affect the way the organism sees

stage 6

• Can add neurons which process signals at different speeds, allows us to detect motion in the outside world 

• Eg light that moves slowly/quickly etc 

• Eventually results in the development of eyes

so what does this tell us about evolution of the eyes

• proceeded from simple non-visual to complex visual eyes 

phototaxis mechanism

• stage 3 is like moths vision

• light shines most strongly on the right eye, so right eye gets the stronger signal, which means left muscle is suddenly activated, begins to twitch 

• If muscle starts twitching, left wing beats harder than it did before, turns in the direction of the light 

• Results in the right eye becoming more strongly activated, wing beats harder, moth will turn further towards the light 

• So moth is compelled to circle the light, cant escape the drive 

how do we think of ourselves as being different

• We like to think of ourselves on not being like that, there arent some signals in the environment that we just have to respond to, we do things because we plan to do them + voluntarily do them

reasons for us not being like a moth

• Our brains are a lot more complex, around 80 billion neurons - level of complexity and computation power is 100 000s magnitude bigger 

We can do anything we want

• How do we want 

• Or is this actually true - do we do things because we want to do them or are there parts of our behaviour that are still automatic, not fully under our control 

• Eg ask to put left hand up, just do it, dont actually know why or how - somehow neural signals from speech transmitted into a pattern of signals into motor cortex that made out hand go up 

• Suggest that most of the things we do are not because we want to do them, but because they happen automatically, we dont know how we do these basic things, eg walking 

• Automatic processes that are the basis of our day to day life are quite moth like

behaviour without awareness in clinical contexts

• visual neglect

• blindsight

what is visual neglect caused by

• damage to the right parietal cortex (e.g., stroke or accident)

charcateritsics of visual neglect

• If not too wide-spread, patient may have no deficit in perception, movement, cognitive skills (eg reasoning), arousal/wakefulness, orientation, motivation, etc.

• But may consistently ignore the left half of the world: read only the right half of a page, wash, dress, shave, or apply make-up to only the right side of the body, eat only what’s on the right side of the plate

• Selective ‘disinterest’ in the left side, but intact ability to make implicit selective choices based on unacknowledged information

formal tests for visual neglect

• line bisection task (explicit)

• object comparison task (explicit)

• implicit knowledge task

line bisection task

• ask pp to cut line in half, neglect: ends up on the far right 

object comparison task

• show a house with fire

• when the fire is on the right, think the pictures are dif

• when the fire is on the left side of the house, think its the same

• so from the studies so far, suggests they just dont process the left side of their world

implicit knowledge task

• dont ask to respond to the feature of interest directly, just ask something which is affected by it  

• Ask which house would prefer to live 

• Patient will reliable pick the one that is not on fire

what does this implicit knowledge task suggest

• Cant be explained if the fire didnt enter their brain, info has to be there somewhere to make the decision

• So can make implicit choice, 

• Info remains entirely unacknowledged 

• Eg when ask why they picked the top one they make up a reason, eg prefer shade of yellow better

• Conclusion: conscious awareness of the left side is impaired, processing of left-side stimuli as such is not impaired. Making sensible decisions, fully processed without ever entering voluntary decision-making

what is blindsight caused by

• Lesion to primary visual cortex causes scotoma (blindness in part of the visual field)

characteristics of blindsight

• If part of the brain that receives visual input is damaged, would assume that whatever corresponds to it in the outside world must be gone - brain can’t process it

• Patients report not seeing anything in that area (subjective blindness) - shown by black spot 

• Patients do not respond to stimuli presented in that area (objective blindness)

localisation task

• Put pp in front of screen and ask to point to the light

• when in the in tact VF, very accurate

• When not in the in tact visual field, don’t point at it, tell them to just guess

• When plot it, close every time, shows it isnt random

• Localisation almost as accurate in the intact visual field 

• But: patients only react to blind-field stimuli when prompted, Insist that they do not see these stimuli

• suggests some residual processing exists

Other examples

• Correct ‘guessing’ of orientation, shape, direction of movement, etc.

in all of these 

• No subjective sense of ‘seeing’ anything

• No spontaneous (unprompted) reaction towards these stimuli

• But spontaneous, voluntary responses (to something else) can be affected by these stimuli:

other study when asked to look at the light

• Task: make an eye-movement (‘saccade’) to a light on the screen (Rafal et al. 1990)

• If ask to make an eye movement in intact VF, does it quickly and accurately

• If show in blind field, pp keeps waiting 

• Present two stimuli simultaneously - reaction time is much longer (doesn’t make sense if  they don’t actually see anything)

• From looking at eye movements, can see that they get sidetracked towards the light 

• So target in blind field not only detected by brain, but processed all the way up to the motor system without this person’s awareness or voluntary control

subliminal perception

• Subliminal = “below the threshold where we can still reliably make an explicit, voluntary judgement” 

• So its about processing things/responding to things that are subjectively invisible 

subliminal perception task

• Task: show an arrow pointing to the left or right

• Ask did the arrow point to the left or to the right?

• Make it more difficult by having the arrow be followed by a masking stimulus 

• Can then make it even harder by having a shorter interval between mask and arrow 

• Good method to measure people's ability to consciously identify a masked stimulus 

• If the mask follows the stimulus quickly enough, the stimulus becomes subjectively invisible

• … but do we still process it?

Subliminal Prime Paradigm method

• make an explicit judgement about the clearly visible stimuli (e.g., did the arrows in stimulus 2 point to the left or to the right?)

• We cannot directly identify stimulus 1, but does it indirectly affect our responses 

• Stimulus 1 points in the same direction = compatible, opposite direction = incompatible, has no direction = neutral (eg two plus signs)

• If stimulus 1 is really invisible (if we don’t process it at all), then its identity shouldn’t make a difference… (should have same RTs for all stimuli) but it does! Makes a systematic difference 

subliminal prime paradigm results

• RTs faster for compatible than incompatible 

• neutral rt is in between the two

• Even though stimulus 1 is subjectively invisible, we still process it - not only processed in visual system, but also affects our voluntary responses to stimulus 2 

hypotheses based off of these results

• Hypothesis 1: Stimulus 1 primes visual processing of stimulus 2 - simply easier to identify the right pointing target 

• But neutral trials, stimuli is also new so should be as slow as the incompatible one

• Hypothesis 2: Stimulus 1 primes its corresponding motor response

• i.e., a left-pointing (subliminal!) arrow causes preparation of a left-hand response, a right-pointing (subliminal!) arrow causes preparation of a right-hand response

• Incompatible - First you need to stop the left response, then prepare the right then execute the right, so more steps in between 

• Also fits with neutral condition - haven’t prepared anything so no response to inhibit 

Subliminal Free Choice Priming Task

• Forced choice - always an arrow pointing left/right, forced to make response to corresponding action of second stimulus 

• Free choice - if see circles, choose to press either arrow. Avoid overarching strategy eg always clicking right 

• If primes improve target identification (hypothesis 1), then responses to the different-looking free-choice targets should not be affected by whether the arrow is pointing to the left or right 

• Prediction: If primes pre-activate their corresponding motor response (hypothesis 2), then motor responses to free-choice targets should be affected 

• following >>, people should be more likely to choose a right-hand than a left-hand response (and should be faster to execute it as it has been pre-activated)

• following <<, people should be more likely to choose a left-hand than a right-hand response (and should be faster to execute it)

free choice results

• forced choice - slower on incompatible

• free choice - people faster when they chose the response that matched the first stimulus. also picked it more often, supports hyp 2

implications from these results

• When people thought they were making a voluntary response, stimulus in environment that systematically shifted responses in one direction or another 

• So there is something we dont see, are not aware that it exists, yet it still affects our choices that we thought were free 

• Subjectively ‘free choices can be subliminally manipulated

Can subliminal stimuli make us do things we don’t want to do (can subliminal control undermine free will) 

• Compared two experimental conditions, one where we had forced choice trials mixed with free choice trials, randomly intermixed - what the previous exp did (found the same pattern)

• Then had a condition without any arrow (forced-choice) targets (i.e., participants only see circle targets & can always choose whether to respond left or right

• Result: The pattern of results above disappears, masked arrow primes no longer affect response choice or speed

• In a neutral env response choices are not subliminally manipulated 

• Conclusion: subliminal arrow primes do not trigger a motor response when participants are not already prepared for them to do so

when do subliminal arrows affect our choices

• When we already want to respond to arrows - already expecting arrow targets 

• In line with our normal response to arrows 

• Manipulation allows you to execute normal responses a bit more quickly and reliably

• Doesn't counteract what you want to do, just in line with it 

fmri

• if neurons are more active, they need more oxygen (they have to ‘breathe harder’)

• Measuring oxygen levels in the brain gives an index of how much activity is going on

• Localizing our measurements tells us where the activity is going on

• Functional Magnetic Resonance Imaging (fMRI) measures oxygen levels through the different magnetic properties of oxygen-rich versus oxygen-poor blood

• Measure where in the brain is there activity that correlates with a task - Shows sig pattern of activity 

• End up with a map of activity (statistical map) - the area that correlated most strongly, didnt actually ‘light up’ 

eegs

• for neurons, being active means exchanging electrical signals

• If sufficiently large clusters of neurons synchronize their electrical activity, the combined electrical signal might become strong enough to be measured even at a distance 

• Electroencephalogram (EEG) uses electrodes on the scalp to measure changes in the electrical currents produced by large groups of synchronized neurons

• If more activity below than above, thingy points down etc

• If do it ms by ms get trace of activity 

erps

• To see if firing patterns reflect specific events, have to repeat event (eg presenting right arrows) 

• Although all look different, is there some signal that remains the same for all right arrows 

• Average after the stimulus is presented, average each data point 

• Although most of the time results around 0 (suggests uncorrelated activity), there is a certain point where it isnt 

• Do firing patterns reflect specific events? YES:

• ERP (Event-Related Potential): the pattern of activity that remains the same for each occurrence of an event

lrps

• Lateralized Readiness Potential (computed from erps)

• In order to map movement activity in the brain, measure LRP

• If left arrow comes up and pp makes left response, means activity in right hem has gone up above baseline (on graph goes down)

• Same pattern true for right hand, at the same time the left hem becomes more active 

• Can see how activity shifts to one side (lateralises) to execute a one handed movement 

• If subtract responses from each other to get rid of noise, get a pattern of activity for compatible response 

• Can see how the brain prepares the movement before it is executed 

• shows how a manual response develops over time

• See how the brain begins to lateralise in the correct direction - how the motor system gears up to execute the response

fmri results from subliminal arrows

• Small area of sig activity in the right motor cortex when there was a left prime 

  - Right prime correlated activity to left motor cortex to navigate left hand response

  
  
  

lrp activity

- Dip down means correct response as executed by the pps 

- if prime was pointed to the right and target pointed to the right, then get blue line 

- Soon after prime is presented, long before target can have any influence, Already see an activation towards the right,

- if the prime was pointing towards the left, in the same time window we see an activation of the side away from the target - activation of the wrong hand 

- so Subliminal prime already triggers motor activity, as predicted by hypothesis two 

- for free choice - if instead of arrow present neutral, hand that corresponds to prime is already activated - biased towards prime