An Extended Normalization Model of Attention Accounts for Feature-Based Attentional Enhancement of Both Response and Coherence Gain

Different gain modulations of selective attention at short and long cue-target delays

Previous studies have tested attentional gain patterns, for example, within the normalization model of attention (NMoA), by altering the relative sizes of the attention field and stimuli. Existing studies have not investigated whether the gain patterns of altering the relative exogenous cue size as compared with the target stimuli matches the prediction of NMoA and whether these gain patterns exist in the late stage of attentional processing. To resolve these questions, the present study maintained the target grating size and changed the exogenous cue size in both short and long cue-target onset asynchronies (CTOAs) conditions. The results revealed response gain for small cue/large target size conditions and contrast gain for large cue/small target size conditions, which was consistent with the NMoA. However, we observed the decrease in the contrast gain factor only with long CTOAs, regardless of whether the cue size was relatively small or large. This indicated that NMoA-related effects based on the relative attentional field dominated in the early stage and that the contrast gain dominated in the late stage.

Temporal attention affects contrast response function by response gain

Orienting attention to a specific point in time has been shown to improve the contrast sensitivity at the attended time point and impair it earlier or later. This phenomenon could be explained by temporal attention increasing the effective contrast of the target presented at the attended time point which leads to changes in contrast psychometric function by contrast gain. Another explanation is that temporal attention multiplicatively amplifies the amplitude of behavioral or neural response to contrast, resulting in alterations in contrast psychometric function by response gain. To explore the underlying mechanism, we adopted a temporal cueing orientation discrimination task using audio pre-cues composed of different frequency components to induce different attentional allocations in the time domain and targets of various contrast intensities to measure contrast psychometric functions. Obtained psychometric functions for contrast sensitivity were fitted for different conditions with discrepant attentional states in time. We found that temporal attention manipulated by cue affected contrast psychometric function by response gain, indicating that multiplying the contrast response of the visual target occurring at the selected point in time by a fixed factor is a crucial way for temporal attention to modulate perceptual processing.

A dynamic normalization model of temporal attention

Vision is dynamic, handling a continuously changing stream of input, yet most models of visual attention are static. Here, we develop a dynamic normalization model of visual temporal attention and constrain it with new psychophysical human data. We manipulated temporal attention-the prioritization of visual information at specific points in time-to a sequence of two stimuli separated by a variable time interval. Voluntary temporal attention improved perceptual sensitivity only over a specific interval range. To explain these data, we modelled voluntary and involuntary attentional gain dynamics. Voluntary gain enhancement took the form of a limited resource over short time intervals, which recovered over time. Taken together, our theoretical and experimental results formalize and generalize the idea of limited attentional resources across space at a single moment to limited resources across time at a single location.

Awareness-Dependent Normalization Framework of Visual Bottom-up Attention

Although bottom-up attention can improve visual performance with and without awareness to the exogenous cue, whether they are governed by a common neural computation remains unclear. Using a modified Posner paradigm with backward masking, we found that the cueing effect displayed a monotonic gradient profile (Gaussian-like), both with and without awareness, whose scope, however, was significantly wider with than without awareness. This awareness-dependent scope offered us a unique opportunity to change the relative size of the attention field to the stimulus, differentially modulating the gain of attentional selection, as proposed by the normalization model of attention. Therefore, for each human subject (male and female), the stimulus size was manipulated as their respective mean attention fields with and without awareness while stimulus contrast was varied in a spatial cueing task. By measuring the gain pattern of contrast-response functions on the spatial cueing effect derived by visible or invisible cues, we observed changes in the cueing effect consonant with changes in contrast gain for visible cues and response gain for invisible cues. Importantly, a complementary analysis confirmed that subjects' awareness-dependent attention fields can be simulated by using the normalization model of attention. Together, our findings indicate an awareness-dependent normalization framework of visual bottom-up attention, placing a necessary constraint, namely, awareness, on our understanding of the neural computations underlying visual attention.SIGNIFICANCE STATEMENT Bottom-up attention is known to improve visual performance with and without awareness. We discovered that manipulating subjects' awareness can modulate their attention fields of visual bottom-up attention, which offers a unique opportunity to regulate its normalization processes. On the one hand, by measuring the gain pattern of contrast-response functions on the spatial cueing effect derived by visible or invisible cues, we observed changes in the cueing effect consonant with changes in contrast gain for visible cues and response gain for invisible cues. On the other hand, by using the normalization model of attention, subjects' awareness-dependent attention fields can be simulated successfully. Our study supports important predictions of the normalization model of visual bottom-up attention and further reveals its dependence on awareness.

To look or not to look: dissociating presaccadic and covert spatial attention

Attention is a central neural process that enables selective and efficient processing of visual information. Individuals can attend to specific visual information either overtly, by making an eye movement to an object of interest, or covertly, without moving their eyes. We review behavioral, neuropsychological, neurophysiological, and computational evidence of presaccadic attentional modulations that occur while preparing saccadic eye movements, and highlight their differences from those of covert spatial endogenous (voluntary) and exogenous (involuntary) attention. We discuss recent studies and experimental procedures on how these different types of attention impact visual performance, alter appearance, differentially modulate the featural representation of basic visual dimensions (orientation and spatial frequency), engage different neural computations, and recruit partially distinct neural substrates. We conclude that presaccadic attention and covert attention are dissociable.

Comparing the influence of stimulus size and contrast on the perception of moving gratings and random dot patterns-A registered report protocol

Modern accounts of visual motion processing in the primate brain emphasize a hierarchy of different regions within the dorsal visual pathway, especially primary visual cortex (V1) and the middle temporal area (MT). However, recent studies have called the idea of a processing pipeline with fixed contributions to motion perception from each area into doubt. Instead, the role that each area plays appears to depend on properties of the stimulus as well as perceptual history. We propose to test this hypothesis in human subjects by comparing motion perception of two commonly used stimulus types: drifting sinusoidal gratings (DSGs) and random dot patterns (RDPs). To avoid potential biases in our approach we are pre-registering our study. We will compare the effects of size and contrast levels on the perception of the direction of motion for DSGs and RDPs. In addition, based on intriguing results in a pilot study, we will also explore the effects of a post-stimulus mask. Our approach will offer valuable insights into how motion is processed by the visual system and guide further behavioral and neurophysiological research.

Different computations underlie overt presaccadic and covert spatial attention

Perception and action are tightly coupled: visual responses at the saccade target are enhanced right before saccade onset. This phenomenon, presaccadic attention, is a form of overt attention-deployment of visual attention with concurrent eye movements. Presaccadic attention is well-documented, but its underlying computational process remains unknown. This is in stark contrast to covert attention-deployment of visual attention without concurrent eye movements-for which the computational processes are well characterized by a normalization model. Here, a series of psychophysical experiments reveal that presaccadic attention modulates visual performance only via response gain changes. A response gain change was observed even when attention field size increased, violating the predictions of a normalization model of attention. Our empirical results and model comparisons reveal that the perceptual modulations by overt presaccadic and covert spatial attention are mediated through different computations.

Effects of visual short-term memory load and attentional demand on the contrast response function

Visual short-term memory (VSTM) load leads to impaired perception during maintenance. Here, we fitted the contrast response function to psychometric orientation discrimination data while also varying attention demand during maintenance to investigate: (1) whether VSTM load effects on perception are mediated by a modulation of the contrast threshold, consistent with contrast gain accounts, or by the function asymptote (1 lapse rate), consistent with response gain accounts; and (2) whether the VSTM load effects on the contrast response function depend on the availability of attentional resources. We manipulated VSTM load via the number of items in the memory set in a color and location VSTM task and assessed the contrast response function for an orientation discrimination task during maintenance. Attention demand was varied through spatial cuing of the orientation stimulus. Higher VSTM load increased the estimated contrast threshold of the contrast response function without affecting the estimated asymptote, but only when the discrimination task demanded attention. When attentional demand was reduced (in the cued conditions), the VSTM load effects on the contrast threshold were eliminated. The results suggest that VSTM load reduces perceptual sensitivity by increasing contrast thresholds, suggestive of a contrast gain modulation mechanism, as long as the perceptual discrimination task demands attention. These findings support recent claims that attentional resources are shared between perception and VSTM maintenance processes.

Flexible top-down modulation in human ventral temporal cortex

Visual neuroscientists have long characterized attention as inducing a scaling or additive effect on fixed parametric functions describing neural responses (e.g., contrast response functions). Here, we instead propose that top-down effects are more complex and manifest in ways that depend not only on attention but also other cognitive processes involved in executing a task. To substantiate this theory, we analyze fMRI responses in human ventral temporal cortex (VTC) in a study where stimulus eccentricity and cognitive task are varied. We find that as stimuli are presented farther into the periphery, bottom-up stimulus-driven responses decline but top-down attentional enhancement increases substantially. This disproportionate enhancement of weak responses cannot be easily explained by conventional models of attention. Furthermore, we find that attentional effects depend on the specific cognitive task performed by the subject, indicating the influence of additional cognitive processes other than attention (e.g., decision-making). The effects we observe replicate in an independent experiment from the same study, and also generalize to a separate study involving different stimulus manipulations (contrast and phase coherence). Our results suggest that a quantitative understanding of top-down modulation requires more nuanced characterization of the multiple cognitive factors involved in completing a perceptual task.

The lateral prefrontal cortex of primates encodes stimulus colors and their behavioral relevance during a match-to-sample task

The lateral prefrontal cortex of primates (lPFC) plays a central role in complex cognitive behavior, in decision-making as well as in guiding top-down attention. However, how and where in lPFC such behaviorally relevant signals are computed is poorly understood. We analyzed neural recordings from chronic microelectrode arrays implanted in lPFC region 8Av/45 of two rhesus macaques. The animals performed a feature match-to-sample task requiring them to match both motion and color information in a test stimulus. This task allowed to separate the encoding of stimulus motion and color from their current behavioral relevance on a trial-by-trial basis. We found that upcoming motor behavior can be robustly predicted from lPFC activity. In addition, we show that 8Av/45 encodes the color of a visual stimulus, regardless of its behavioral relevance. Most notably, whether a color matches the searched-for color can be decoded independent of a trial's motor outcome and while subjects detect unique feature conjunctions of color and motion. Thus, macaque area 8Av/45 computes, among other task-relevant information, the behavioral relevance of visual color features. Such a signal is most critical for both the selection of responses as well as the deployment of top-down modulatory signals, like feature-based attention.

Normalization governs attentional modulation within human visual cortex

Although attention is known to increase the gain of visuocortical responses, its underlying neural computations remain unclear. Here, we use fMRI to test the hypothesis that a neural population’s ability to be modulated by attention is dependent on divisive normalization. To do so, we leverage the feature-tuned properties of normalization and find that visuocortical responses to stimuli sharing features normalize each other more strongly. Comparing these normalization measures to measures of attentional modulation, we demonstrate that subpopulations which exhibit stronger normalization also exhibit larger attentional benefits. In a converging experiment, we reveal that attentional benefits are greatest when a subpopulation is forced into a state of stronger normalization. Taken together, these results suggest that the degree to which a subpopulation exhibits normalization plays a role in dictating its potential for attentional benefits.

Attention is known to enhance relevant information in our environment, yet its underlying neural computations remain unclear. Here, the authors provide evidence that the degree to which a neural population can normalize itself results in greater potential for attentional benefits.

Strategic deployment of feature-based attentional gain in primate visual cortex

Attending to visual stimuli enhances the gain of those neurons in primate visual cortex that preferentially respond to the matching locations and features (on-target gain). Although this is well suited to enhance the neuronal representation of attended stimuli, it is nonoptimal under difficult discrimination conditions, as in the presence of similar distractors. In such cases, directing attention to neighboring neuronal populations (off-target gain) has been shown to be the most efficient strategy, but although such a strategic deployment of attention has been shown behaviorally, its underlying neural mechanisms are unknown. Here, we investigated how attention affects the population responses of neurons in the middle temporal (MT) visual area of rhesus monkeys to bidirectional movement inside the neurons' receptive field (RF). The monkeys were trained to focus their attention onto the fixation spot or to detect a direction or speed change in one of the motion directions (the "target"), ignoring the distractor motion. Population activity profiles were determined by systematically varying the patterns' directions while maintaining a constant angle between them. As expected, the response profiles show a peak for each of the 2 motion directions. Switching spatial attention from the fixation spot into the RF enhanced the peak representing the attended stimulus and suppressed the distractor representation. Importantly, the population data show a direction-dependent attentional modulation that does not peak at the target feature but rather along the slopes of the activity profile representing the target direction. Our results show that attentional gains are strategically deployed to optimize the discriminability of target stimuli, in line with an optimal gain mechanism proposed by Navalpakkam and Itti.

Electrical stimulation of macaque lateral prefrontal cortex modulates oculomotor behavior indicative of a disruption of top-down attention

The lateral prefrontal cortex (lPFC) of primates is hypothesized to be heavily involved in decision-making and selective visual attention. Recent neurophysiological evidence suggests that information necessary for an orchestration of those high-level cognitive factors are indeed represented in the lPFC. However, we know little about the specific contribution of sub-networks within lPFC to the deployment of top-down influences that can be measured in extrastriate visual cortex. Here, we systematically applied electrical stimulations to areas 8Av and 45 of two macaque monkeys performing a concurrent goal-directed saccade task. Despite using currents well above saccadic thresholds of the directly adjacent Frontal Eye Fields (FEF), saccades were only rarely evoked by the stimulation. Instead, two types of behavioral effects were observed: Stimulations of caudal sites in 8Av (close to FEF) shortened or prolonged saccadic reaction times, depending on the task-instructed saccade, while rostral stimulations of 8Av/45 seem to affect the relative attentional weighting of saccade targets as well as saccadic reaction times. These results illuminate important differences in the causal involvement of different sub-networks within the lPFC and are most compatible with a stimulation-induced biasing of stimulus processing that accelerates the detection of saccade targets presented ipsilateral to stimulation through a disruption of contralaterally deployed top-down attention.