Attention as a decision in information space

Neural dynamics of attentional cross-modality control

Attentional networks that integrate many cortical and subcortical elements dynamically control mental processes to focus on specific events and make a decision. The resources of attentional processing are finite. Nevertheless, we often face situations in which it is necessary to simultaneously process several modalities, for example, to switch attention between players in a soccer field. Here we use a global brain mode description to build a model of attentional control dynamics. This model is based on sequential information processing stability conditions that are realized through nonsymmetric inhibition in cortical circuits. In particular, we analyze the dynamics of attentional switching and focus in the case of parallel processing of three interacting mental modalities. Using an excitatory-inhibitory network, we investigate how the bifurcations between different attentional control strategies depend on the stimuli and analyze the relationship between the time of attention focus and the strength of the stimuli. We discuss the interplay between attention and decision-making: in this context, a decision-making process is a controllable bifurcation of the attention strategy. We also suggest the dynamical evaluation of attentional resources in neural sequence processing.

Gearing up for action: attentive tracking dynamically tunes sensory and motor oscillations in the alpha and beta band

Allocation of attention during goal-directed behavior entails simultaneous processing of relevant and attenuation of irrelevant information. How the brain delegates such processes when confronted with dynamic (biological motion) stimuli and harnesses relevant sensory information for sculpting prospective responses remains unclear. We analyzed neuromagnetic signals that were recorded while participants attentively tracked an actor's pointing movement that ended at the location where subsequently the response-cue indicated the required response. We found the observers' spatial allocation of attention to be dynamically reflected in lateralized parieto-occipital alpha (8–12 Hz) activity and to have a lasting influence on motor preparation. Specifically, beta (16–25 Hz) power modulation reflected observers' tendency to selectively prepare for a spatially compatible response even before knowing the required one. We discuss the observed frequency-specific and temporally evolving neural activity within a framework of integrated visuomotor processing and point towards possible implications about the mechanisms involved in action observation.

•

Observing dynamic actions modulates on-going alpha and beta neural activity.

•

Alpha modulations reflect dynamic changes in the allocation of spatial attention.

•

Beta modulations relate to evolving, stimulus location-based response bias.

•

Alpha and beta activity jointly contribute to sensorimotor integration for action.

Thinking in spatial terms: decoupling spatial representation from sensorimotor control in monkey posterior parietal areas 7a and LIP

Perhaps the simplest and most complete description of the cerebral cortex is that it is a sensorimotor controller whose primary purpose is to represent stimuli and movements, and adaptively control the mapping between them. However, in order to think, the cerebral cortex has to generate patterns of neuronal activity that encode abstract, generalized information independently of ongoing sensorimotor events. A critical question confronting cognitive systems neuroscience at present therefore is how neural signals encoding abstract information emerge within the sensorimotor control networks of the brain. In this review, we approach that question in the context of the neural representation of space in posterior parietal cortex of non-human primates. We describe evidence indicating that parietal cortex generates a hierarchy of spatial representations with three basic levels: including (1) sensorimotor signals that are tightly coupled to stimuli or movements, (2) sensorimotor signals modified in strength or timing to mediate cognition (examples include attention, working memory, and decision-processing), as well as (3) signals that encode frankly abstract spatial information (such as spatial relationships or categories) generalizing across a wide diversity of specific stimulus conditions. Here we summarize the evidence for this hierarchy, and consider data showing that signals at higher levels derive from signals at lower levels. That in turn could help characterize neural mechanisms that derive a capacity for abstraction from sensorimotor experience.

The role of uncertainty and reward on eye movements in a virtual driving task

Eye movements during natural tasks are well coordinated with ongoing task demands and many variables could influence gaze strategies. Sprague and Ballard (2003) proposed a gaze-scheduling model that uses a utility-weighted uncertainty metric to prioritize fixations on task-relevant objects and predicted that human gaze should be influenced by both reward structure and task-relevant uncertainties. To test this conjecture, we tracked the eye movements of participants in a simulated driving task where uncertainty and implicit reward (via task priority) were varied. Participants were instructed to simultaneously perform a Follow Task where they followed a lead car at a specific distance and a Speed Task where they drove at an exact speed. We varied implicit reward by instructing the participants to emphasize one task over the other and varied uncertainty in the Speed Task with the presence or absence of uniform noise added to the car's velocity. Subjects' gaze data were classified for the image content near fixation and segmented into looks. Gaze measures, including look proportion, duration and interlook interval, showed that drivers more closely monitor the speedometer if it had a high level of uncertainty, but only if it was also associated with high task priority or implicit reward. The interaction observed appears to be an example of a simple mechanism whereby the reduction of visual uncertainty is gated by behavioral relevance. This lends qualitative support for the primary variables controlling gaze allocation proposed in the Sprague and Ballard model.

Twenty-five lessons from computational neuromodulation

Neural processing faces three rather different, and perniciously tied, communication problems. First, computation is radically distributed, yet point-to-point interconnections are limited. Second, the bulk of these connections are semantically uniform, lacking differentiation at their targets that could tag particular sorts of information. Third, the brain's structure is relatively fixed, and yet different sorts of input, forms of processing, and rules for determining the output are appropriate under different, and possibly rapidly changing, conditions. Neuromodulators address these problems by their multifarious and broad distribution, by enjoying specialized receptor types in partially specific anatomical arrangements, and by their ability to mold the activity and sensitivity of neurons and the strength and plasticity of their synapses. Here, I offer a computationally focused review of algorithmic and implementational motifs associated with neuromodulators, using decision making in the face of uncertainty as a running example.

Attention, learning, and the value of information

Despite many studies on selective attention, fundamental questions remain about its nature and neural mechanisms. Here I draw from the animal and machine learning fields that describe attention as a mechanism for active learning and uncertainty reduction and explore the implications of this view for understanding visual attention and eye movement control. I propose that a closer integration of these different views has the potential greatly to expand our understanding of oculomotor control and our ability to use this system as a window into high level but poorly understood cognitive functions, including the capacity for curiosity and exploration and for inferring internal models of the external world.

Cognitive functions of the posterior parietal cortex: top-down and bottom-up attentional control

Although much less is known about human parietal cortex than that of homologous monkey cortex, recent studies, employing neuroimaging, and neuropsychological methods, have begun to elucidate increasingly fine-grained functional and structural distinctions. This review is focused on recent neuroimaging and neuropsychological studies elucidating the cognitive roles of dorsal and ventral regions of parietal cortex in top-down and bottom-up attentional orienting, and on the interaction between the two attentional allocation mechanisms. Evidence is reviewed arguing that regions along the dorsal areas of the parietal cortex, including the superior parietal lobule (SPL) are involved in top-down attentional orienting, while ventral regions including the temporo-parietal junction (TPJ) are involved in bottom-up attentional orienting.

The attentional drift-diffusion model extends to simple purchasing decisions

How do we make simple purchasing decisions (e.g., whether or not to buy a product at a given price)? Previous work has shown that the attentional drift-diffusion model (aDDM) can provide accurate quantitative descriptions of the psychometric data for binary and trinary value-based choices, and of how the choice process is guided by visual attention. Here we extend the aDDM to the case of purchasing decisions, and test it using an eye-tracking experiment. We find that the model also provides a reasonably accurate quantitative description of the relationship between choice, reaction time, and visual fixations using parameters that are very similar to those that best fit the previous data. The only critical difference is that the choice biases induced by the fixations are about half as big in purchasing decisions as in binary choices. This suggests that a similar computational process is used to make binary choices, trinary choices, and simple purchasing decisions.

Building Bridges between Perceptual and Economic Decision-Making: Neural and Computational Mechanisms

Investigation into the neural and computational bases of decision-making has proceeded in two parallel but distinct streams. Perceptual decision-making (PDM) is concerned with how observers detect, discriminate, and categorize noisy sensory information. Economic decision-making (EDM) explores how options are selected on the basis of their reinforcement history. Traditionally, the sub-fields of PDM and EDM have employed different paradigms, proposed different mechanistic models, explored different brain regions, disagreed about whether decisions approach optimality. Nevertheless, we argue that there is a common framework for understanding decisions made in both tasks, under which an agent has to combine sensory information (what is the stimulus) with value information (what is it worth). We review computational models of the decision process typically used in PDM, based around the idea that decisions involve a serial integration of evidence, and assess their applicability to decisions between good and gambles. Subsequently, we consider the contribution of three key brain regions – the parietal cortex, the basal ganglia, and the orbitofrontal cortex (OFC) – to perceptual and EDM, with a focus on the mechanisms by which sensory and reward information are integrated during choice. We find that although the parietal cortex is often implicated in the integration of sensory evidence, there is evidence for its role in encoding the expected value of a decision. Similarly, although much research has emphasized the role of the striatum and OFC in value-guided choices, they may play an important role in categorization of perceptual information. In conclusion, we consider how findings from the two fields might be brought together, in order to move toward a general framework for understanding decision-making in humans and other primates.

Unique and shared roles of the posterior parietal and dorsolateral prefrontal cortex in cognitive functions

The dorsolateral prefrontal cortex (PFC) and posterior parietal cortex (PPC) are two parts of a broader brain network involved in the control of cognitive functions such as working-memory, spatial attention, and decision-making. The two areas share many functional properties and exhibit similar patterns of activation during the execution of mental operations. However, neurophysiological experiments in non-human primates have also documented subtle differences, revealing functional specialization within the fronto-parietal network. These differences include the ability of the PFC to influence memory performance, attention allocation, and motor responses to a greater extent, and to resist interference by distracting stimuli. In recent years, distinct cellular and anatomical differences have been identified, offering insights into how functional specialization is achieved. This article reviews the common functions and functional differences between the PFC and PPC, and their underlying mechanisms.

Inhibition of return in a visual foraging task in non-human subjects

Inhibition of return is thought to help guide visual search by inhibiting the orienting of attention to previously attended locations. We have previously shown that, in a foraging visual search task, the neural responses to objects in parietal cortex are reduced after they have been examined. Here we ask whether the animals' reaction times (RTs) in the same task show a psychophysical correlate of inhibition of return: a slowing of reaction time in response to a probe placed at a previously fixated location. We trained three animals to perform an RT version of the visual foraging task. In the foraging task, subjects visually searched through an array of five identical distractors and five identical potential targets; one of which had a reward linked to it. In the RT variant of the task, subjects had to rapidly respond to a probe if it appeared. We found that RTs were slower for probes presented at locations that contained previously fixated objects, faster to potential targets and between the two for behaviorally irrelevant distractors that had not been fixated. These data show behavioral inhibitory tagging of previously fixated objects and suggest that the suppression of activity seen previously in the same task in parietal cortex could be a neural correlate of this mechanism.

Specific contributions of ventromedial, anterior cingulate, and lateral prefrontal cortex for attentional selection and stimulus valuation

Functional clusters of neurons in the monkey prefrontal and anterior cingulate cortex are involved in guiding attention to the most valuable objects in a scene.

Attentional control ensures that neuronal processes prioritize the most relevant stimulus in a given environment. Controlling which stimulus is attended thus originates from neurons encoding the relevance of stimuli, i.e. their expected value, in hand with neurons encoding contextual information about stimulus locations, features, and rules that guide the conditional allocation of attention. Here, we examined how these distinct processes are encoded and integrated in macaque prefrontal cortex (PFC) by mapping their functional topographies at the time of attentional stimulus selection. We find confined clusters of neurons in ventromedial PFC (vmPFC) that predominantly convey stimulus valuation information during attention shifts. These valuation signals were topographically largely separated from neurons predicting the stimulus location to which attention covertly shifted, and which were evident across the complete medial-to-lateral extent of the PFC, encompassing anterior cingulate cortex (ACC), and lateral PFC (LPFC). LPFC responses showed particularly early-onset selectivity and primarily facilitated attention shifts to contralateral targets. Spatial selectivity within ACC was delayed and heterogeneous, with similar proportions of facilitated and suppressed responses during contralateral attention shifts. The integration of spatial and valuation signals about attentional target stimuli was observed in a confined cluster of neurons at the intersection of vmPFC, ACC, and LPFC. These results suggest that valuation processes reflecting stimulus-specific outcome predictions are recruited during covert attentional control. Value predictions and the spatial identification of attentional targets were conveyed by largely separate neuronal populations, but were integrated locally at the intersection of three major prefrontal areas, which may constitute a functional hub within the larger attentional control network.

To navigate within an environment filled with sensory stimuli, the brain must selectively process only the most relevant sensory information. Identifying and shifting attention to the most relevant sensory stimulus requires integrating information about its sensory features as well as its relative value, that is, whether it's worth noticing. In this study, we describe groups of neurons in the monkey prefrontal cortex that convey signals relating to the value of a stimulus and its defining feature and location at the very moment when attention is shifted to the stimulus. We found that signals conveying information about value were clustered in a ventromedial prefrontal region, and were separated from sensory signals within the anterior cingulate cortex and the lateral prefrontal cortex. The integration of valuation and other “top-down” processes, however, was achieved by neurons clustered at the intersection of ventromedial, anterior cingulate, and lateral prefrontal cortex. We conclude that valuation processes are recruited when attention is shifted, independent of any overt behavior. Moreover, our analysis suggests that valuation processes can bias the initiation of attention shifts, as well as ensure sustained attentional focusing.

Deficits in attentional control: cholinergic mechanisms and circuitry-based treatment approaches

The cognitive control of attention involves maintaining task rules in working memory (or "online"), monitoring reward and error rates, filtering distractors, and suppressing prepotent, and competitive responses. Weak attentional control increases distractibility and causes attentional lapses, impulsivity, and attentional fatigue. Levels of tonic cholinergic activity (changes over tens of seconds or minutes) modulate cortical circuitry as a function of the demands on cognitive control. Increased cholinergic modulation enhances the representation of cues, by augmenting cue-evoked activity in thalamic glutamatergic afferents, thereby increasing the rate of detection. Such cholinergic modulation is mediated primarily via α4β2* nicotinic acetylcholine receptors. Animal experiments and clinical trials in adult patients with ADHD indicate that attentional symptoms and disorders may benefit from drugs that stimulate this receptor. Tonic cholinergic modulation of cue-evoked glutamatergic transients in prefrontal regions is an essential component of the brain's executive circuitry. This circuitry model guides the development of treatments of deficits in attentional control.

Mechanisms for generating and compensating for the smallest possible saccades

Microsaccades are small eye movements that occur during gaze fixation. Although taking place only when we attempt to stabilize gaze position, microsaccades can be understood by relating them to the larger voluntary saccades, which abruptly shift gaze position. Starting from this approach to microsaccade analysis, I show how it can lead to significant insight about the generation and functional role of these eye movements. Like larger saccades, microsaccades are now known to be generated by brainstem structures involved not only in compiling motor commands for eye movements, but also in identifying and selecting salient target locations in the visual environment. In addition, these small eye movements both influence and are influenced by sensory and cognitive processes in various areas of the brain, and in a manner that is similar to the interactions between larger saccades and sensory or cognitive processes. By approaching the study of microsaccades from the perspective of what has been learned about their larger counterparts, we are now in a position to make greater strides in our understanding of the function of the smallest possible saccadic eye movements.

Neural mechanisms of saccade target selection: gated accumulator model of the visual-motor cascade

We review a new computational model developed to understand how evidence about stimulus salience in visual search is translated into a saccade command. The model uses the activity of visually responsive neurons in the frontal eye field as evidence for stimulus salience that is accumulated in a network of stochastic accumulators to produce accurate and timely saccades. We discovered that only when the input to the accumulation process was gated could the model account for the variability in search performance and predict the dynamics of movement neuron discharge rates. This union of cognitive modeling and neurophysiology indicates how the visual-motor transformation can occur, and provides a concrete mapping between neuron function and specific cognitive processes.

Human prosaccades and antisaccades under risk: effects of penalties and rewards on visual selection and the value of actions

Monkey studies report greater activity in the lateral intraparietal area and more efficient saccades when targets coincide with the location of prior reward cues, even when cue location does not indicate which responses will be rewarded. This suggests that reward can modulate spatial attention and visual selection independent of the "action value" of the motor response. Our goal was first to determine whether reward modulated visual selection similarly in humans, and next, to discover whether reward and penalty differed in effect, if cue effects were greater for cognitively demanding antisaccades, and if financial consequences that were contingent on stimulus location had spatially selective effects. We found that motivational cues reduced all latencies, more for reward than penalty. There was an "inhibition-of-return"-like effect at the location of the cue, but unlike the results in monkeys, cue valence did not modify this effect in prosaccades, and the inhibition-of-return effect was slightly increased rather than decreased in antisaccades. When financial consequences were contingent on target location, locations without reward or penalty consequences lost the benefits seen in noncontingent trials, whereas locations with consequences maintained their gains. We conclude that unlike monkeys, humans show reward effects not on visual selection but on the value of actions. The human saccadic system has both the capacity to enhance responses to multiple locations simultaneously, and the flexibility to focus motivational enhancement only on locations with financial consequences. Reward is more effective than penalty, and both interact with the additional attentional demands of the antisaccade task.

Eye movements: the past 25 years

This article reviews the past 25 years of research on eye movements (1986-2011). Emphasis is on three oculomotor behaviors: gaze control, smooth pursuit and saccades, and on their interactions with vision. Focus over the past 25 years has remained on the fundamental and classical questions: What are the mechanisms that keep gaze stable with either stationary or moving targets? How does the motion of the image on the retina affect vision? Where do we look - and why - when performing a complex task? How can the world appear clear and stable despite continual movements of the eyes? The past 25 years of investigation of these questions has seen progress and transformations at all levels due to new approaches (behavioral, neural and theoretical) aimed at studying how eye movements cope with real-world visual and cognitive demands. The work has led to a better understanding of how prediction, learning and attention work with sensory signals to contribute to the effective operation of eye movements in visually rich environments.

Frontal-to-parietal top-down causal streams along the dorsal attention network exclusively mediate voluntary orienting of attention

Previous effective connectivity analyses of functional magnetic resonance imaging (fMRI) have revealed dynamic causal streams along the dorsal attention network (DAN) during voluntary attentional control in the human brain. During resting state, however, fMRI has shown that the DAN is also intrinsically configured by functional connectivity, even in the absence of explicit task demands, and that may conflict with effective connectivity studies. To resolve this contradiction, we performed an effective connectivity analysis based on partial Granger causality (pGC) on event-related fMRI data during Posner's cueing paradigm while optimizing experimental and imaging parameters for pGC analysis. Analysis by pGC can factor out exogenous or latent influences due to unmeasured variables. Typical regions along the DAN with greater activation during orienting than withholding of attention were selected as regions of interest (ROIs). pGC analysis on fMRI data from the ROIs showed that frontal-to-parietal top-down causal streams along the DAN appeared during (voluntary) orienting, but not during other, less-attentive and/or resting-like conditions. These results demonstrate that these causal streams along the DAN exclusively mediate voluntary covert orienting. These findings suggest that neural representations of attention in frontal regions are at the top of the hierarchy of the DAN for embodying voluntary attentional control.

Variance as a signature of neural computations during decision making

Traditionally, insights into neural computation have been furnished by averaged firing rates from many stimulus repetitions or trials. We pursue an analysis of neural response variance to unveil neural computations that cannot be discerned from measures of average firing rate. We analyzed single-neuron recordings from the lateral intraparietal area (LIP), during a perceptual decision-making task. Spike count variance was divided into two components using the law of total variance for doubly stochastic processes: (1) variance of counts that would be produced by a stochastic point process with a given rate, and loosely (2) the variance of the rates that would produce those counts (i.e., "conditional expectation"). The variance and correlation of the conditional expectation exposed several neural mechanisms: mixtures of firing rate states preceding the decision, accumulation of stochastic "evidence" during decision formation, and a stereotyped response at decision end. These analyses help to differentiate among several alternative decision-making models.

Predicting the timing of wrong decisions with LATER

Response time, or latency, is increasingly being used to provide information about neural decision processes. LATER (Linear Approach to Threshold with Ergodic Rate) is a quasi-Bayesian model of decision-making, with the additional feature that it introduces a degree of gratuitous randomisation into the decision process. It has had some success in predicting latencies under various conditions, but has not specifically been applied to an equally important aspect of decision-making, namely errors: a complete model of decision-making should not only account for latency distributions of correct decisions but also of wrong ones. We therefore used a decision task that generates large numbers of errors: subjects are told to look at suddenly appearing targets of one colour, but not another. We found that subjects' faster responses are as likely to be correct as wrong, but eventually the latency distributions diverge, with errors becoming infrequent. It seems that colour information, arriving after a delay, results both in cancellation of the developing response to the mere existence of the target and in delayed initiation of the correct response. A simple model, using LATER units in a similar way to one that has previously successfully modelled countermanding, accurately predicts latency distributions and proportions of all responses, whether correct or incorrect, demonstrating that the LATER model can indeed account for errors as well as correct responses.

The involvement of posterior parietal cortex and frontal eye fields in spatially primed visual search

BACKGROUND

Right posterior parietal cortex (rPPC) and frontal eye fields (FEF) are known to be involved in processing visuospatial attention. However, the functional involvement of these areas in spatial priming in complex conjunction visual search has yet to be determined.

OBJECTIVE

This study aimed to examine the roles of rPPC and bilateral FEF in conjunction search when spatial ambiguity was reduced by priming the target location.

METHODS

Participants completed a conjunction search task whereby the target location was random or else repeated from the previous trial. Transcranial magnetic stimulation was delivered to each one of the three sites of interest at a time, and task performance was compared with a sham condition.

RESULTS

Spatial priming occurred for all conditions: search times were faster for primed relative to nonprimed trials. When the target appeared at a nonprimed location, stimulation over any of the three sites increased reaction times relative to the sham condition. However, when the target location was repeated, reaction time was only significantly increased by stimulation over the right FEF.

CONCLUSIONS

rPPC and left FEF are only involved when the target location is random, suggesting that these areas are essential for resolving spatial ambiguity to localize targets. Conversely, right FEF contributes equally to visual search regardless of spatial priming. We propose that right FEF has a role in the integration of bottom up saliency and top down expectancy signals and is the node at which rPPC and/or left FEF is either recruited or not.

Spatial and non-spatial functions of the parietal cortex

Although the parietal cortex is traditionally associated with spatial attention and sensorimotor integration, recent evidence also implicates it in higher order cognitive functions. We review relevant results from neuron recording studies showing that inferior parietal neurons integrate information regarding target location with a variety of non-spatial signals. Some of these signals are modulatory and alter a stimulus-evoked response according to the action, category, or reward associated with the stimulus. Other non-spatial inputs act independently, encoding the context or rules of a task even before the presentation of a specific target. Despite the ubiquity of non-spatial information in individual neurons, reversible inactivation of the parietal lobe affects only spatial orienting of attention and gaze, but not non-spatial aspects of performance. This suggests that non-spatial signals contribute to an underlying spatial computation, possibly allowing the brain to determine which targets are worthy of attention or action in a given task context.