Show Tag: attention

Select Other Tags

There is a theory called the `premotor theory of visual attention' which posits that activity that can ultimately lead to a saccade can also facilitate processing of stimuli in those places the saccade will/would go to.

A possible ascending pathway from SC to visual cortex through the pulvinar nuclei (pulvinar) may be responsible for the effect of SC activity on visual processing in the cortex.

Feature-based and spatial attention may be based on similar mechanisms.

Feature-based visual attention facilitates object detection across the visual field.

The effects of visual attention are more pronounced in later stages of visual processing (in the visual cortex).

Spatial attention does not seem to affect the selectivity of visual neurons—just the vigour of their response.

Spatial visual attention increases the activity of neurons in the visual cortex whose receptive fields overlap the attended region.

Feature-based visual attention increases the activity of neurons in the visual cortex which respond to the attended feature.

Spatial and feature-based visual attention are additive: together, they particularly enhance the activity of any neuron whose receptive field encompasses the attended region, contains a stimulus with the attended feature, and prefers that feature.

Humans can learn to use the statistics of their environment to guide their visual attention.

Humans do not need to be aware of the stimulus they perceive to use them to guide their visual attention.

Stimuli in one modality can guide attention in another.

Humans can learn to use stimuli in one modality to guide attention in another.

Palmer and Ramsey show that lack of awareness of a visual lip stream does not inhibit learning of its relevance for a visual localization task: the subliminal lip stream influences visual attention and affects the subjects' performance.

They also showed that similar subliminal lip streams did not affect the occurrence of the Mc Gurk effect.

Together, this suggests that awareness of a visual stimulus is not always needed to use it for guiding visual awareness, but sometimes it is needed for multisensory integration to occur (following Palmer and Ramsey's definition).

There are parallels between visual attention and eye movements because both serve the purpose of directing our processing of visual information to stimuli from a region in space that is small enough to handle for our brain.

Since visual attention and eye movements are so tightly connected in the process of visual exploration of a scene, it has been suggested that the same mechanisms may be (partially) responsible for guiding them.

There is evidence suggesting that one cannot plan a saccade to one point in space and turn covert visual attention to another at the same time.

It has been found that stimulating supposed motor neurons in the SC facilitates visual processing in the part of visual cortex whose receptive field is the same as that of the SC stimulated neurons.

Feature-based visual attention facilitates neural responses across the visual field (in visual cortex).

Born et al. provided evidence which shows that preparing a saccade alone already enhances visual processing at the target of the saccade: discrimination targets presented before saccade onset were identified more successfully if they were in the location of the saccade target than when they were not.

Born et al. showed that, if the color of a saccade target stimulus is task relevant, then identification of a discrimination target with that same color is enhanced even if it is not in the same location.

Search targets which share few features with mutually similar distractors surrounding them are said to `pop out': it seems to require hardly any effort to identify them and search for them is very fast.

Search targets that share most features with their surrounding, on the other hand, require much more time time be identified.

Saccades evoked by electric stimulation of the deep SC can be deviated towards the target of visual spatial attention. This is the case even if the task forbids a saccade towards the target of visual spatial attention.

Activation build-up in build-up neurons is modulated by spatial attention.

Kustov's and Robinson's results support the hypothesis that there is a strong connection of action and attention.

Botvinick et al. advance two interdependent hypotheses:

  1. Conflicts in information processing activate certain cortical areas, most notably the anterior cingulate cortex,
  2. Conflict-related activity causes adjustments in cognitive control of information processing to resolve conflict.

Spatial attention raises baseline activity in neurons whose RF are where the attention is even without a visual stimulus (in visual cortex).

Unilateral lesions in brain areas associated with attention can lead to visuospatial neglect; the failure to consider anything within a certain region of the visual field. In extreme cases this can mean that patients e.g. only read from one side of a book.

Kastner and Ungerleider propose that the top-down signals which lead to the effects of visual attention originate from brain regions outside the visual cortex.

Regions lesions of which can induce visuospatial neglect include

  • the parietal lobe, in particular the inferior part,
  • temporo-parietal junction,
  • the anterior cingulate cortex,
  • basal ganglia,
  • thalamus,
  • the pulvinar nucleus.

Many of the cortical areas projecting to the SC have been implicated with attention.

Dehner et al. speculate that the inhibitory influence of FAES activity on SIV activity is connected to modality-specific attention: According to that hypothesis, an auditory stimulus which leads to strong FAES activity will suppress activity in FAES and thus block out cortical somatosensory input to the SC.

Spatial attention can enhance the activity of SC neurons whose receptive fields overlap the attended region

AES has been implicated with selective attention.

Schenck summarizes three neurorobotic studies in which he evaluates visual prediction, and, more specifically, predictive remapping. He argues that his experiments support a claim in psychology saying that pre-saccadic activation of neurons whose receptive fields will contain the location of a salient stimulus after the saccade is not just pre-activation but actually a prediction of what the visual field will be like after the saccade.

Sprague and Meikle Jr. propose that the SC is involved in visual attention.

There are neurons in the supplementary eye field which are related to

  • eye movements,
  • arm movements,
  • ear movements,
  • spatial attention.

Attention is necessary to perform the Stroop and Simon tasks.

The frontoparietal network seems involved in executive control and orienting.

The anterior cingulate cortex is likely involved with regulating attention.

Attention developed quite early. Even very simple organisms, like drosophila and honeybees, show evidence of attentional processes.

It has been found that stimulating supposed motor neurons in the SC enhances responses of v4 neurons with the same receptive field as the SC neurons.

Krauzlis et al. state that collicular deactivation has not been found to eliminate signs of task-based attention in neural responses in cortex.

Krauzlis et al. argue that SC deactivation should have changed neural responses in cortex if it regulated attention through visual cortex.

Krauzlis et al.'s argument that SC deactivation should have changed neural responses in cortex if it regulated attention through visual cortex is a bit weak considering that stimulating SC does change sensory representations in v4.

Krauzlis et al. argue that animals without a well-developed neocortex nonetheless show signs of visual attention. Thus, it is likely that the neocortex is not necessary for attention and SC can regulate attention without the neocortex.

Krauzlis et al. argument that animals without a well-developed neocortex show signs of selective attention similar to humans and other higher animals shows that neocortex may not be necessary for attention seems more appropriate than that of lack of influence of collicular deactivation on cortical responses.

Krauzlis et al. argue that attention may not so much be a explicit mechanism but a phenomenon emerging from the need of distributed information processing systems (biological and artificial) for centralized coordination:

According to that view, some centralized control estimates the state of (some part of) the world and modulates both action and perception according to the state which is estimated to be the most plausible at any given point.

Krauzlis et al. localize this central control in the basal ganglia.

VIsual attention is the facilitation of visual processing of some stimuli over others.

``The heminanopia that follows unilateral removal of the cortex that mediates visual behavior cannot be explained simply in classical terms of interruption of the visual behavior cannot be explained simply in classical terms of interruption of the visual radiations that serve cortical function.
Explanation fo the deficit requires a broader point of view, namely, that visual attention and perception are mediated at both forebrain and midbrain levels, which interact in their control of visually guided behavior.''

(Sprague, 1966)

Anderson suggests that it would make sense if we attended to whatever to attend to promises the greatest reward.

Saliency of a stimulus might say something about its likelihood of offering reward if attended to.

The probability of reward of attending a stimuli is influenced by two factors:

  • the probability of selecting the right thing,
    Thus, highly distinctive things have great bottom-up saliency.
  • the probability of reward given that the right thing is selected,
    • Features that are associated with high reward salient
    • Goals can affect which features promise reward in a situation.

Visual feature combinations become more salient if they are learned to be associated with reward.

Targets which are selected in one trial tend to be more salient in subsequent trials—they are selected faster and rejected slower.

The extent of this effect is modulated by whether or not the selection was rewarded.

Verschure summarizes version VII of his distributed adaptive control model as "a unifying theory" of perception cognition, and action. He states that it uses a learned world model in its contextual layer which biases perception processing (top-down) on the one hand, and saliency (bottom-up) on the other. Between these to appears to be what he calls the validation gate which defines matching and mismatch between world model and percepts.

If there are a number of stimuli, many of which share the same low-level features, then that stimulus that does not "pops out". Local saliency-based models like the one due to Itty and Koch fail to explain this effect.

Task-irrelevant cues in one modality can enhance reaction times in others—but they don't always do that. Instances of this effect have been implicated with exogenous attention.

Task-irrelevant auditory cues have been found to enhance reaction times in others. visual cues, however, which cued visual localization, did not cue auditory localization.

Fixating some point in space enhances spoken language understanding if the words come from that point in space. Fixating a visual stream showing lips consistent with the utterances, this effect is strongest, but it also works if the visual display is random. The effect is also enhanced if fixation is combined with some form of visual task which is complex enough.

Fixating at some point in space can impede language understanding if the utterance do not emanate from the focus of visual attention and there are auditory distractors which do.

Goldberg and Wurtz found that neurons in the superficial SC respond more vigorously to visual stimuli in their receptive field if the current task is to make a saccade to the stimuli.

Responses of superficial SC neurons do not depend solely to intrinsic stimulus properties.

Traditionally, visual attention is subdivided into feature-based attention and spatial attention. However, spatial is arguably only one cue out of possibly a number of cues and possibly only a special case.

Weber and Triesch's model learns task-relevant features.

However, a brain region like the SC, which serves a very general task, cannot specialize in one task—it has to serve all goals that the system has.

It therefore should change its behavior depending on the task. Attention is one mechanism which might determine how to change behavior in a given situation.

If the goal is predictive of the input, then a purely unsupervised algorithm could take a representation of the goal as just another input.

While it is possible that the goal often is predictive of the input, some error feedback is probably necessary to tune the degree to which the algorithm can be `distracted' by task-irrelevant but interesting stimuli.

There are voluntary (endogenous) and reflexive (exogenous) mechanisms of guiding selective attention.

Santangelo and Macaluso describe typical experiments for studying visual attention.

Frontal eye fields (FEF) and intraparietal sulcus (IPS) have been associated with voluntary orienting of visual attention.

Santangelo and Macaluso provide a rewiew on the recent literature on visual and auditory attention.

Frontoparietal regions play a key role in spatial orienting in unisensory studies of visual and auditory attention.

There seems to be also modality-specific attention which globally de-activates attention in one modality and activates it in the other.

As a computer scientist I would call de-activating one modality completely a special case of selective attention in that modality.

Localized auditory cues can exogenously orient visual attention.

Santangelo and Macaluso state that multisensory integration and attention are probably separate processes.

Maybe attention controls whether or not multi-sensory integration (MSI) happens at all (at least in SC)? That would be in line with findings that without input from AES and rLS, there's no MSI.

Are AES and rLS cat homologues to the regions cited by Santangelo and Macalluso as regions responsible for auditory and visual attention?

Task-irrelevant visual cues do not affect visual orienting (visual spatial attention). Task-irrelevant auditory cues, however, seem to do so.

Santangelo and Macaluso suggest that whether or not the effects of endogenous attention dominate the ones of bottom-up processing (automatic processing) depends on semantic association, be it linguistic or learned association (like dogs and barking, cows and mooing).

Santangelo and Macaluso state that "the same frontoparietal attention control systems are ... activated in spatial orienting tasks for both the visual and auditory modality..."

De Kamps and van der Velde argue for combinatorial productivity and systematicity as fundamental concepts for cognitive representations. They introduce a neural blackboard architecture which implements these principles for visual processing and in particular for object-based attention.

De Kamps and van der Velde use their blackboard architecture for two very different tasks: representing sentence structure and object attention.

Deco and Rolls introduce a system that uses a trace learning rule to learn recognition of more and more complex visual features in successive layers of a neural architecture. In each layer, the specificity of the features increases together with the receptive fields of neurons until the receptive fields span most of the visual range and the features actually code for objects. This model thus is a model of the development of object-based attention.

Using multiple layers each of which learns with a trace rule with successively larger time scales is similar to the CTRNNs Stefan Heinrich uses to learn the structure of language. Could there be a combined model of learning of sentence structure and language processing on the one hand and object-based visual or multi-modal attention on the other?

One function, or, what Santangelo and macaluso call `the key element', of selective attention is filtering out distracters—ie. noise filtering.

Stimuli which are non-predictive in a task—like localized stimuli in one modality which are non-predictive of the position of the target in another modality—can enhance performance in valid instances of that task—like detecting targets which by coincidence are where the non-predictive stimulus was.

This demonstrates the existence of exogenous attention.

When asked to ignore stimuli in the visual modality and attend to the auditory modality, increased activity in the auditory temporal cortex and decreased activity in the visual occipital cortex can be observed (and vice versa).

Bertelson et al. did not find a shift of sound source localization due to manipulated endogenous visual spatial attention—localization was shifted only due to (the salience of) light flashes which would induce (automatic, mandatory) exogenous attention.

A localized visual stimulus can shorten the response to a target stimulus if it appears near and shortly after the first stimulus.

It can lengthen the response time if the target stimulus appears somewhere else or too late.

A localized visual stimulus can lengthen the response time to a target if the target stimulus appears somewhere too late after the first stimulus.

This is called `inhibition of return'.

Bell et al. found that playing a sound before a visual target stimulus did not increase activity in the neurons they monitored for long enough to lead to (neuron-level) multisensory integration.

Bell et al. make it sound like enhancement in SC neurons due to exogenous, visual, spatial attention is due to residual cue-related activity which is combined (non-linearly) with target-related activity.

If enhancement in SC neurons due to exogenous, visual, spatial attention is due to residual cue-related activity which is combined (non-linearly) with target-related activity, then that casts an interesting light on (the lack of) intra-modal enhancement:

The only difference between an intra-modal cue-stimulus combination and an intra-modal stimulus-stimulus combination lies in the temporal order of the two. Therefore, if two visual stimuli presented in the receptive field of an SC at the same time) neuron do not enhance the response to each other, then the reason can only be a matter of timing.

In an fMRI experiment, Fairhall and Macaluso found that attending (endogenously, spatially) to congruent audio-visual stimuli (moving lips and speech) produced greater activation in SC than either attending to non-congruent stimuli or not attending to congruent stimuli.

Visual spatial attention

  • lowers the stimulus detection threshold,
  • improves stimulus discrimination,

With two stimuli in the receptive field, one with features of a visual search target and one with different features

  • increases average neural activity in cortex compared to the same two objects without attending to any features
  • decreases average neural activity if spatial attention is on the location of the non-target compared to when it is on the target.

The fact that average neural activity in cortex is decreased if spatial attention is on the location of a non-target out of a target and a non-target compared to when it is on the target supports the notion that inhibition plays an important role in stimulus selection.

Divisive normalization models describe neural responses well in cases of

  • olfactory perception in drosophila,
  • visual processing in retina and V1,
  • possibly in other cortical areas,
  • modulation of responses through attention in visual cortex.

Common metaphors for attention are the 'spotlight' metaphor, the 'bottleneck' metaphor and the 'zoomlens' metaphor. At the core of these metaphors is the notion that attention is a mechanism which regulates to which information some limited resource is applied.

Attention affects both early and late perceptual processing.

Divisive normalization models have explained how attention can facilitate or suppress some neurons' responses.

Some models view attentional changes of neural responses as the result of Bayesian inference about the world based on changing priors.

Chalk et al. argue that changing the task should not change expectations—change the prior—about the state of the world. Rather, they might change the model of how reward depends on the state of the world.

Before a saccade is made, the region that will be the target of that saccade is perceived with higher contrast and visual contrast.

People fixate on different parts of an image depending on the questions they are asked or task they are trying to accomplish.

Brouwer et al found that their subjects looked more at the contact position of the index finger when they were told to grasp an object than when they were just to look at it.

In the first experiment by Brouwer et al, people fixated different parts of a shape depending on whether the task was just to look at it or grasp it.

The subject's initial saccade, however, was not influenced by the task.

It makes sense that sub-cortical visual processing uses peripheral information more than cortical processing:

  • sub-cortical processing is concerned with latent monitoring of the environment for potential dangers (or conspecifiics)
  • sub-cortical processing is concerned with watching the environment and guiding attention in cortical processing.

Grossberg states that ART predicts a functional link between consciousness, learning, expectation, attention, resonance, and synchrony and calls this principle the CLEARS principle.

The changes to neural responses due to top-down attention are purely caused by intrinsic processes, not a (direct) reaction to external stimuli. They thus support the theory of situatedness.

LIP has been suggested to contain a saliency map of the visual field, to guide visual attention, and to decide about saccades.

Saccade targets tend to be the centers of objects.

Lateral prefrontal cortex (LPFC) and frontal eye fields (FEF) are frontal cortex regions involved in visual attention and target selection.

Both populations in prefrontal cortex and posterior parietal cortex show correlates of bottom-up and top-down visual attention.

In the pop-out condition of a visual search task, Buschman and Miller found that neurons in the posterior parietal cortex region LIP found the search target earlier than neurons in frontal cortex regions FEF and LPFC.

In the pure visual search condition of a visual search task, Buschman and Miller found that neurons in frontal cortex regions FEF and LPFC found the search target earlier than neurons in the posterior parietal cortex region LIP.

Visual attention is influenced both by local and global saliency, ie. bottom-up processes, and by semantics, ie. top-down processes.

The biased competition theory of visual attention explains attention as the effect of low-level stimuli competing with each other for resources—representation and processing. According to this theory, higher-level processes/brain regions bias this competition.

Predictive coding and biased competition are closely related concepts. Spratling combines them in his model and uses it to explain visual saliency.

Desimone and Duncan argue that spatial information about a search target can be part of the attentional template fitted against all potential targets in the visual display as any other object feature.