Consciousness tweede deeltentamen
College 9 & 10:
Attention, free will and consciousness.
Change blindness: miss something because you are not
playing attention to it.
What is attention for?
From low level features to high level features and then in
the motor cortex more or less the other way around.
→ the obvious problem is you cannot translate any
sensory input into motor output.
→ So which sensory inputs are selected for action?
That’s the prime function of attentions. IN many paradigms this can be
seen. A simple dichotic listening experiment with shadowing shows the
limits of the perception action cycle. Subjects hear two different auditory
streams in two ears. They can only reproduce and remember one stream
at a time: the stream they attend to. This is limited capacity in going
from perception to action.
Both on cognitive level but also mechanically (you simply cannot say two
outputs at the same time).
So this is the core function of attention: SELECTION.
In the visual domain, attending is either achieved via over ot covert attention.
1. Overt orienting and attending: moving your eyes, body, ears, nose, etc in the direction
of a relevant stimulus (what the superior collicus does).
2. Covert attention: shifting your attention towards something, without any external ,
overt signs, i.e., while maintaining fixations. This mechanism has evolved particularly
in social animals, in which direction of gaze often has a strong meaning (threat,
aggression, sexual attention).
Covert attention also exists in the auditory domain (cocktail party effect).
→ one may pretend to listen to someone in front of you while actually focusing on what is
said in another conversation.
Or listening to the person in front of you, your attention may be captured by someone saying
your name in another conversation = attentional capture.
Types of attention:
1. Top down attention
2. Capture/ bottom up
3. Object based attention
4. Feature based attention.
,Top down attention: when subjects are
instructed to focus their attention on some
location of the visual field (such as the Posner
cueing task). The behavioral effect typically is
that the reaction times to presented targets are
faster at the attended location.
So in an invalid trial people are slower, because
they are instructed to attend at the other side.
Bottom up attention: a suddenly appearing
stimulus will automatically ‘capture’ attention.
Shorter reaction time to primed location. This
happens even when the subjects know the prime
is mostly invalid.
Subjects are told to ignore the prime. This
experiment shows that capture is automatic and
not top down.
Attentional capture transforms into inhibition of return when the temporal interval between
the prime and target is over 300 ms. Now, the reaction time is longer for the cued location.
The subject starts to actively suppress attention to the location of the (which was mostly
invalid) cue.
Above here, all sorts of attentions are forms of spatial attention.
Object based attention: instead of focusing on a particular location,
attention can also focus on a particular object.
How to show the difference: cueing C leads to faster detection of S than of
D, even when there is an occlude overlying the bars. Attention can be
directed towards objects that overlap in space.
Issue here: are these really non-spatial forms of attention? Or is that
spatial attention simply spreading fasting along objects, and can spatial
attention also work in 3d?
Another clever experiment to show the existence of object based attention:
one of two white lines was cued, and then targets where sown on either the
same location (valid), on the same white line (invalid within) or the other
white line (invalid between).
A difference in RT between invalid targets on the two lines was only
perceived when they perceived as different objects, not when perceived as
a single background surface, proving the existence of object based
attention.
,Feature based attention: shorter reaction times to (objects with) features that are
attended.
What is the effect of attention on processing?
In top down attention, the effect of enhanced response to stimuli that appear at the attended
location. The posner cueing task increases the amplitutde of visual evoked potentials recorded
from the human scalp (P1).
→ this enhanced response is specific for the attended location (so it is not general or overall
increase in neural responsiveness). The P1 attention effect is topographically distributed over
the visual cortex in accordance with the location of the visual field where attention is directed
at.
So this is different than for example arousal, where there is an overall effect. Attention is very
location specific.
Attentional capture has a similar effect on neural
processing: an increased response for stimuli at the cued
(or rather captured) location. Inhibition of return results
in a decreased response at the cued location, in
accordance with the now decreased attention at that site.
(on the right you see the effect of inhibition of return).
This effect of capture (and inhibition of return) again
show spatial specificity , in that they occur only at the
cortical locations where the attention is directed to.
In monkey area V4, there is an increased response when
attention is directed towards the preferred stimulus
(Desimone). Spatial attention causes increased signal.
Attention to a feature (direction of motion) enhances response to that
feature all over the visual field. Feature based attention causes increased
signals of attended features.
On the left: transparent motion (two sets of moving dots moving in
opposite directions). Monkey attends one of these. Records from cells
responding to motion on the right.
→ responses are stronger (positive attentional modulation) when motion
direction on the right is the same as the direction of motion that is attending
on the left.
, Feature based attention (colour) in area V4. Fixation dot reveals which color to attend (and
later discriminate). Record activity from task-irrelevant stimuli.
Object based attention increases signals of the attended object:
Attending to faces enhances activity in the FFA, while attention to houses enhances activity in
the parahippocampal place area.
In the monkey area V4, attention increases the amount of
synchrony between neurons that process the attended stimulus.
→ activity is recorded from two separate neurons with distinct
receptive fields (yellow and green boxes). When the grating
stimulus that covers these RF’s is attended, responses are larger
than when the other grating is attended.
→ But also, the strength of synchronous 50Hz oscillations
between the two neurons is larger when the stimulus covering the
RF’s is attended.
So not only more action potentials but also more synchronized.
We have seen synchrony as the potential label for assembly coding: neurons of each assembly
fire action potentials in synchrony. Assembly A and B code for different objects. The brain
‘knows’ which parts belong together because those belong to an assembly fire action
potentials in synchrony (cross-correlation function reveals synchrony).
Increased synchrony and increased response
have both been proposed as a neural correlate
of perceptual grouping/binding, needed for
object recognition.
Does the increased synchronyand reponse
strength seen when the stimuli are attended
play a role in (better) object recognition?
1. Feature integration theory
2. Biased competition theory
Feature integration theory → visual search task (find wally).
Wally is uniquely defined by a combination of different features.
