Neuronal Attention (6b) Flashcards
(14 cards)
Briefly describe the original evidence for attentional competition in temporal cortex
MONKEY NEUROPHYSIOLOGY provided some of the original evidence for how attentional competition is resolved at the neuronal level
–> FOUND individual neurons in monkey temporal cortex show preferences
they respond SELECTIVELY to a particular type of stimulus
Monkey Task - Chelazzi et al. (1993)
Chelazzi trained monkeys to make an eye movement towards a target
–> They were CUED which stimulus to respond to
eg. if they saw rectangle cue, they had to make an eye movement towards the rectangle
They FOUND:
–> Initially: individual neurons responded equally strongly regardless of whether the preferred or nonpreferred stimulus was the target
–> After 180ms: if preferred is target then firing remains high, if non-preferred stimulus is the target the response of the neuron is suppressed
CLEARLY SHOWS: that neuronal responses in inferior temporal cortex are competitive
–> demonstrates that visual attention operates independently of the eyes, and before a response is carried out
What three things does this evidence (Chelazzi) tell us about attention
- Attentional competition occurs at the level of INDIVIDUAL neurons
–> involves both EXCITATION (preference) and INHIBITION (no preference) - MODULATION of neuronal responses by attention occurs well BEFORE response occurs
- Competition occurs in the brain regions that process the visual features of relevant
–> the same neurons that process the visual features of an object
How early in the processing stream do such attentional modulation effects occur: outline fMRI research mapping
Attentional effects on early visual processing have also been found with fMRI
–> shows modulation of a variety of different cortical regions by attention
Research: Brefcynski and DeYoe (1999)
Different regions of a circular area were cues and subjects had to make judgements on stimuli that subsequently appeared in that region
Displayed Segment
–> researchers presented visual stimuli in different segments of a circle individually, and used RETINOTOPIC MAPPING to examine which areas of visual cortex were sensitive to visual stimulation in different regions of space
Cued Segment
–> subjects auditory cued to attend to a specific segment
–> subjects then had to judge the colour and orientation of lines in a specific segment of the array
FOUND:
- same pattern of activation regardless of whether something presented in segment, or being orientated towards that segment
- can create exact same spatial map of activation by just getting participants to attend to certain areas of space
ERP and fMRI studies converge to show attentional modulation of cortical signals in primary visual cortex
–> there is an ANALOGUE MAP in V1 corresponding to attentional effects in the circular processing field
Other regions: Kastner (1998) Attentional Competition Effects in V4
fMRI evidence for attentional modulation of neuronal signals in V4
–> coloured patterns were presented either sequentially or simultaneously to subjects in the MRI scanner
First Tested: whether stimuli compete for attention
–> FOUND: when items presented sequentially, responses in visual cortex were higher than when they were presented simultaneoudly
*activation in V4 is lower when presented simultanously
Second Tested: instructed ppts to attend to one of the objects
- this effect disappeared
–> response to object attending to was equally high as it was when presented alone
–> resolution of the attentional competition effect
*again demonstrates that attention modulates competitive interactions at the neuronal level in the visual cortex
–> competitive interactions in the visual and temporal cortex play a role in attentional selection
O’Craven et al. (1999): Attentional modulation in FFA and PPA
fMRI evidence for attentional modulation of neuronal signals in FFA and PPA
reminder FFA prefers faces, PPA prefers houses
What they did:
- subjects presented with overlapping pictures of faces and houses in the MRI scanner
- face or house moved
- subjects attended to either moving or static object
What they found:
- activation in FFA and PPA depended on which stimulus was being attended to
- when asking to attend to the MOVING OBJECT
–> if the moving object is preferred (faces) the FFA activation is elevated
–> if the moving object is not preferred (houses) the FFA activation is suppressed
AGAIN demonstrated that attention modulates competitive interactions at the neuronal level in the ventral visual pathway
–> attention effectively filters out sensory input that is irrelevant to current task demands
–> by enhancing neuronal responses in regions processing task-relevant inputs
–> and by suppressing responses in regions processing task-irrelevant inputs
*this modulation is both in early processing (eg. V1) and late processing (eg. FFA)
–> attentional system can operate flexibly at different stages depending on task demands
Reminder: Nillie Lavie - Load Theory (what did they do/find)
What they did
- varied the perceptual load (high or low) - either having a variety of distractors or all the same distractors
- and the congruency between the distractor and target (could either be same letter or different letter)
- calculated the difference in reaction time when the distractor was incongruent relative to when it was congruent in each of the conditions
Found:
–> - incongruent distractor has much more of an effect in the EASIER (low load) condition
–> HIGH perceptual load, your capacity is used up by processing the targets and non-targets, so there is none left over for distractors (early selection)
–> LOW perceptual load, you’re not using up all your resources, and there’s some left to process distractors (late selection)
Schwartz et al. 2005 - neural correlations of perceptual load effects
Attentional modulation of visual cortical signals depends on perceptual load
Had subjects perform two conditions
1. Low load - detect any red shape
2. High load - detect any specific conjunction of shape or colour (eg. yellow upright)
Main task: was flanked by checkerboard stimuli which produce high levels of activation in visual cortex
Graph shows the difference in activation between high and low load conditions
–> Visual cortex activation due to checkerboard stimuli is MUCH HIGHER in the low load condition
–> Essentially, in the high load condition subjects are so focused on the main task they can filter out the irrelevant checkerboard stimuli
–> earliest level of processing these were observed
Summarise the findings of fMRI research
Neuroimaging studies show attentional modulation of neuronal signalling in sensory-specific brain regions
Such modulation occurs throughout the sensory pathways - suggesting attention is a FLEXIBLE system for resolving competition for awareness at multiple different levels
HOWEVER: fMRI provides correlation data, can’t infer causation
Taylor et al. 2006 - EEG Evidence for visual cortex activation
- what did they do
- what did they find
Evidence for top-down bias signals originating in frontoparietal cortex - the effects of TMS over Frontal Eye Fields on activation in visual cortex
This study provided evidence for a CAUSAL ROLE OF PFC ACTIVATION ON ATTENTIONAL SIGNALS IN VISUAL CORTEX
What:
- subjects performed a posner cueing task in which a central cue pointed left or right, and a target then appeared on the left or right
- used TMS to stimulate the frontal eye fields between the cue and the target
Found:
- Subjects are faster to respond to validly cued targets than invalidly cued targets
- responses were slower during TMS of the frontal eye fields (much more slowing in the invalid condition) when eye field is stimulated compared to stimulating a control area
- the normal attention-related negativity seen with no TMS is MARKEDLY REDUCED IN THE TMS CONDITION
–> demonstrates that signals over the PFC have a causal effect on signals in VC during attention
These findings suggest that attentional selection may involve LONG-DISTANCE INTERACTIONS between PFC and Visual Cortex
Underlying mechanism of interaction: outline the idea of neuronal frequency synchronisation
Current hypothesis is that during attentional tasks, the PFC sends biasing signals to sensory-specific regions of the cortex - enhancing task-relevant activation, and suppressing task-irrelevant region
Neuronal Frequency Synchronisation: different brain regions can fall into frequency, idea is that when there is synchronisation - the brain regions are working together
- there are 5 frequency bands of EEG signals (that oscillate at particular frequencies)
- different frequencies might correspond to different function
Outline the work by Buschman and Miller (2007) - synchronisation
IN this study the researchers got monkeys to perform a visual search task involving either easy or hard search
–> they examined: the extent to which regions in the parietal and PF cortex sociallated at the same frequency
FOUND:
- coherence was higher in the middle frequency band during conjunction (hard search)
- but coherence was higher in the upper frequency band during popout search (easy search)
Provides the explanation that: top-down, voluntary attention depends on frequency synchronisation between parietal and prefrontal cortex in the MIDDLE BAND
Whilst bottom-up, reflexive attention depends on fr. synch. in the UPPER BAND
explains how distant, remote regions on the brain can work together
Summarise the lecture (competition, modulation effect, PFC, coherence)
- competition for selection is evident at multiple neuronal levels in brain regions that have evolved to perform other function (eg. sensory processing)
- however, modulation of sensory cortical signals during attention MAY be a downstream effect of attentional signals coming from elsewhere
- the PFC plays a key role as the source of such top down modulation, in the form of a BIASING SIGNAL leading to this enhancement/suppression in lower level sensory-specific brain regions
- coherence (frequency synchronisation) seems to provide a mechanism whereby prefrontal cortex can communicate with posterior cortical regions, to ENABLE attentional selection
