Steven J. Luck - US grants

This project will examine the cognitive and neural mechanisms of visual selective attention in humans. Several experiments will assess the role that attention plays when human subjects perform visual discriminations, ranging from the detection of simple features such as color and orientation to the discrimination of complex conjunctions of form and color. These experiments will determine whether spatially focused attention is necessary for performing these discriminations, and the outcome of these experiments will have important implications for the computational role of selective attention. Additional experiments will explore the specific neural structures in which selective attention operates and the time-course of attentional allocation. The allocation of attention will be measured with a combination of behavioral measurements and electrophysiological recordings (using the event-related potential technique). The necessity of spatially focused attention for a given discrimination will be assessed by requiring subjects to perform two concurrent tasks: 1) a primary task that requires the subject to perform the discrimination of interest; and 2) a secondary task that will absorb any available attentional resources and ensure that subjects do not focus attention onto the primary task targets unless these targets cannot be discriminated without spatially focused attention. The neural structures in which attention operates will be assessed by measurements of electrophysiological responses with known neural generator sources. The time-course of attentional allocation will be measured by using behavioral and electrophysiological measures of attentional allocation to track the focusing of attention over time as subjects scan stimulus arrays. This research is relevant for several important mental health issues. Specifically, several psychiatric and neurological disturbances such as schizophrenia are characterized by deficits in attention, and an understanding of the function of attention in normal individuals is important for understanding and ameliorating these disorders. In addition, many developmental disorders such as dyslexia and attention deficit disorder also involve deficits in attentional processes. Reading, in particular, is extremely dependent on the mechanisms of attention that will be studied in this project, and the results of the proposed experiments may ultimately be useful for developing treatments for reading disabilities.

Although the human brain is an incredibly powerful computational device, it does not have sufficient capacity to process the thousands of sights, sounds, memories, thoughts, and emotions that may arise at any given moment. Mechanisms of selective attention are therefore used to focus the brain's processing resources onto a subset of the many possible sources of information. Previously, attention has been considered to be a unitary cognitive process, but recent evidence has suggested that separable attention mechanisms operate at different stages of processing. The goal of the present project is to isolate and characterize these multiple attention mechanisms, focusing on the stages of visual perception, short-term memory, and response selection. Because these separate attention mechanisms may concurrently influence overt behavior, it is difficult to isolate them by means of traditional behavioral measures of performance. We will therefore use electrophysiological recordings of the brain's activity (colloquially known as `brain waves` and technically as `event-related potentials`) to measure the influence of selective attention at different stages of processing. Although selective processing may be necessary at different stages of processing and may require different neural mechanisms, the brain must also maintain unity of behavioral output, and we will therefore also study how the different attention mechanisms are linked together into a coherent system. This research will help us to understand how the brain copes with an overload of input sources, which will have important implications for the design of future computing devices, and for remediating disorders of cognition.

This project will investigate a fundamental aspect of human perception, the ability to organize the visual world into figures and backgrounds. In everyday visual scenes, objects overlap and occlude one another, which necessitates a visual process for determining which objects lie in the foreground (figures) and which lie in the background. Psychologists have proposed a set of rules that describe figure-ground segregation. For example, a symmetrical region is more likely to be perceived as a figure than as the background. However, this descriptive approach does not specify the mechanisms of figure-ground segregation-the precise computations that allow these rules to be instantiated. Neurophysiological studies in nonhuman primates, in contrast, show promise for elucidating the mechanisms of figure-ground segregation, but the results of these studies are difficult to link with studies of human perception. The goal of this project is to bridge the gap between human perception and monkey neuro-physiology in figure-ground segregation by using noninvasive electro-physiological recordings of the brain's activity (colloquially known as 'brain waves' and technically as 'event-related potentials'). These recordings will allow us to assess the time course and neuroanatomical substrates of figure-ground segregation, thereby elucidating both the cognitive and neural mechanisms of figure-ground segregation. We hypothesize that figure-ground segregation relies on a mechanism that amplifies the neural representation of a figure relative to the neural representation of the ground. Our research uses non-invasive electro-physiological methods that can test this hypothesis more directly by measuring the sensory processes associated with both "figures" and "grounds." This research will provide an understanding of how the brain processes complex visual scenes, which may have implications for artificial vision systems and techniques for rehabilitation of visual perception following brain injury.

This project will examine the operation of attention at two coarsely defined stages of processing, namely visual perception and visual working memory. Most theories of attention do not accommodate the possibility that attention might have different properties in different cognitive subsystems, and the goal of this project will be to demonstrate that there are actually several differences between the attentional mechanisms that operate during perception and those that operate during working memory. For example, we will test the hypothesis that the highly spatiotopic organization of visual perception is mirrored by perceptual- level attentional mechanisms with spatiotopic properties, such as a broad spatial gradient and an inability to focus attention on two separate locations without attending to the intervening region. This contrasts with the more abstract nature of object representations in working memory, which we expect to be mirrored by attentional mechanisms that will lack the spatiotopic properties of perceptual-level attention and instead be object-based. To examine the properties of perceptual-level and working memory-level attention, we will use a combination of behavioral/psychophysical methods and electrophysiological recordings (specifically event-related potentials). In particular, we will compare "memory-intensive" tasks in which working memory is overloaded but the perceptual demands are minimal with "perception-intensive" tasks in which memory is not overloaded but the perceptual demands are great. Attention should operate at different stages in these tasks, allowing us to isolate the properties of perceptual-level and working memory-level attentional mechanisms. This program of research will have important long-term implications for psychological/psychiatric disorders in which attention is compromised, such as attention deficit disorder, reading disorders, and schizophrenia. Specifically, by developing methods to isolate specific attentional mechanisms and by assessing the characteristics of each attentional mechanism, it will be easier to identify the specific attentional mechanisms that are compromised in a given disorder and to understand how the attentional impairments impact the overall disorder.

Abstract
Human activity and thought is embedded within and richly structured by the space around us. We have detailed knowledge of the objects that surround us-where they are, what they are, how they are arranged relative to one another. And we can easily remember the layout of objects in completely new situations, quickly learning which things go where. This grant investigates the dynamics of spatial cognition, that is, the time-dependent processes that underlie such coordinated spatial behaviors.
Spatial cognition has typically been studied as isolated parts-spatial perception, spatial memory, spatial attention, and so on. Consistent with this approach, neurophysiological evidence suggests a functional and anatomical segregation of the visual system into one neural pathway that represents spatial locations ("where") and another neural pathway that represents object property information ("what"). Critically, we know very little about how information in these two pathways is integrated, despite the fact that most behaviors rely on both spatial and object information.
The goal of this grant is to tackle the integration of "where" and "what" systems using a neurally-plausible theory of working memory, the Dynamic Neural Field Theory (DNFT). This theory captures how people hold information in working memory, how they use perceptual cues to keep memorized information accurate, and how long-term memories emerge from this mix. Critically, our new model specifies not just where objects are located, but what those objects are and how spatial and object information can be brought together to guide action. The research plan formalizes this new model and tests a set of critical predictions across 10 experiments.
The DNFT is the first theory of spatial cognition that integrates perception, working memory, and long-term memory in a neurally-plausible way that makes specific predictions about how people behave. As such, this project will advance our understanding of the processes that govern human activity in space-how people think about space, how people organize spatial activities, and the local "maps" of the world people bring with them from context to context. Such information could have a critical impact on how groups and individuals structure spaces to foster communication and lessen memory and attentional demands. Moreover, there is compelling evidence that deficits in where-what integration underlie the behavior problems prevalent in several mental health disorders. The DNFT takes an important step toward addressing both the behavioral and neural processes that likely underlie these behavioral disorders. The space around us has a profound effect on human thought and activity. The proposed research will move us closer to understanding this pervasive aspect of human and social dynamics.
The present work will advance two educational goals as well. First, the PI and co-PIs will establish an international graduate student exchange to foster the education of promising scholars in both the US and Germany. Second, funds from this grant will support a high school outreach program to expose AP calculus students to the application of mathematical models in "real world" science. Beyond these educational impacts, the proposed project will impact the general scholarly community through the publication of empirical and theoretical manuscripts and the creation of a web-based dynamic field simulator.

[unreadable] DESCRIPTION (provided by applicant): Working memory plays a central role in cognitive processing, and an understanding of working memory is essential if we are to understand simple cognitive tasks such as retrieving a fact from memory, complex cognitive tasks such as cooking a meal, and psychopathologies that involve impaired cognition such as schizophrenia. The proposed research seeks to further our understanding of the visual storage component of the working memory system by means of a combination of traditional cognitive methods, psychophysical procedures adapted from vision research, computational simulations of explicit models, and human electrophysiological recordings (ERPs). This project will focus on four fundamental questions about the nature of visual working memory representations and the processes that create those representations. First, we will explore how the multiple features of an object are bound together in working memory, testing the hypothesis that bound objects are the fundamental units of visual working memory storage. Second, we will determine whether visual working memory consists of a small set of discrete, fixed-resolution representations or a potentially large set of variable-resolution representations. Third, we will ask whether working memory representations consist of structured sets of basic features or activations of potentially rich long-term memory representations (or both). Finally, we will examine the processes that transform transient and fragile perceptual representations into durable working memory representations that can survive the passage time and the perception of new stimuli. These issues will be addressed at the cognitive level, but they will also be linked to an explicit hypothesis about the neural mechanisms of working memory storage. The general goal of this project is to provide a deeper understanding of the most fundamental aspects of visual working memory, pushing basic science research on working memory to the next level. At the same time, the methods and findings of this project will provide the backbone for our laboratory's ongoing program of translational research on schizophrenia. Impairments in working memory are a key component of a spectrum of severe and pervasive cognitive dysfunctions in schizophrenia; these treatment-refractory cognitive impairments are largely responsible for the functional disability that is characteristic of the illness. However, it has been difficult to precisely characterize the nature of the working memory impairment in schizophrenia with the psychometric methods usually used in clinical research. This makes it difficult to define specific cognitive targets for new treatments or to provide a precise assessment of the effectiveness of these treatments. The methods and findings that will follow from the proposed research will make it possible to more accurately characterize and measure working memory deficits, which is an important component of ongoing efforts to develop treatments for cognitive impairments in schizophrenia. [unreadable] [unreadable]

DESCRIPTION (provided by applicant): The purpose of this proposed R25 NIH Research Education Grant is to continue supporting a highly successful ten-day intensive summer training workshop in the event-related potential (ERP) technique. This technique is widely used to measure the time course of brain activity across a range of basic science and clinical science domains. The workshop is intended for beginning and intermediate ERP researchers, and it will provide them with training in the essential aspects of the ERP technique, giving them a solid foundation for conducting research on normal and abnormal functioning of the mind and brain. The project director is a well-established researcher and educator who has experience addressing both basic science and clinical science questions with the ERP technique. The workshops will be held at the Center for Mind & Brain at UC-Davis, the home of a large and vibrant community of ERP researchers. In addition to the project director, eleven faculty from UC-Davis and three faculty from other universities will serve as the faculty for the yearly workshops. The workshop participants will come from a variety of disciplines, including cognitive science, cognitive neuroscience, clinical psychology, experimental psychopathology, neurology, psychiatry, developmental psychology, gerontology, speech pathology, and reading disorders. This multi-disciplinary group of participants reflects the broad use of ERPs across the many domains that focus on neural and mental functioning. The participants will also represent a variety of career stages, including graduate and medical students, postdocs, residents, junior faculty, and senior faculty, which reflects the fact that the ERP technique is now being adopted by people at all career stages. The workshop will involve a combination of lectures, discussions, and laboratory activities. An integrated set of lectures on the fundamentals of ERP research will be given by the project director. Lectures on special topics, including advanced techniques and applications to clinical populations, will be given by the other faculty. Several small-group discussions will also be included, led by the workshop faculty, that will focus on the methods used in previously published ERP papers. A series of guided lab activities will also be included. Significant outreach and dissemination efforts will spread the educational benefits of this workshop to a broader group of researchers. These efforts are based on a web site, http://erpinfo.org, which provides lecture notes, information about methods, data analysis software, and sample datasets to the worldwide research community. The ultimate goal of this project is to increase the quality and quantity of future ERP research so that this important technique can have a greater impact on scientific progress in the many basic science and clinical disciplines that focus on the human mind and brain.

Project Summary The goal of our research program is to identify specific, low-level ?building blocks? of cognition that are impaired in people with schizophrenia (PSZ), are linked with neurobiology, and can explain deficits in a higher- level cognitive function. Our prior research has led to the hyperfocusing hypothesis, which proposes that many aspects of cognitive impairment in PSZ can be traced to overly narrow and intense focusing of processing resources on a subset of available inputs, and an inability to distribute resources among multiple sources of information. Hyperfocusing can explain reduced working memory capacity and impaired performance in a variety of attention tasks, and measures of hyperfocusing are strongly correlated with measures of broad cognitive function (e.g., IQ). We have developed a working model of the underlying neurobiology, which predicts that hyperfocusing will lead to exaggerated competitive inhibition and increased repulsion between similar neural representations, and we have obtained preliminary evidence showing that PSZ exhibit the predicted increase in repulsion between representations. If hyperfocusing can be established as a significant mechanism underlying cognitive dysfunction in PSZ, this will set the stage for (a) the near-term development of cognitive training treatments that are designed to counteract hyperfocusing, and (b) the medium-term development of biotherapies that target the neurobiology underlying hyperfocusing. However, additional research is necessary before moving to treatment development. In Aim 1, we will use several new behavioral paradigms to extend the scope of the hyperfocusing hypothesis. In each of these tasks, the hyperfocusing hypothesis predicts that PSZ will exhibit supranormal attention effects, which cannot easily be explained by generalized deficits. In Aim 2, we will use state-of-the-art multivariate fMRI and EEG-based methods to provide more direct evidence of narrower but more intense neural activity in PSZ. These methods will make it possible to go beyond measuring simple activity levels and assess the neural representations formed by PSZ and control subjects. In Aim 3, we will use newly developed fMRI and ERP methods to test the hypothesis that hyperfocusing leads to increased attention-driven competitive inhibition in PSZ. Each of the experiments for Aims 1-3 will involve medium sample sizes (40 PSZ and 40 controls). For Aim 4, each participant across the 5-year project period will also be tested in a set of 7 core tasks that are hypothesized to reflect either hyperfocusing or cognitive control, along with measures of broad cognitive function and functional outcome/capacity. This will provide data from 200+ subjects that will allow us to test the hypothesis that hyperfocusing and cognitive control are separate factors that explain unique variance in cognitive ability and functional outcome/capacity. Together, these 4 aims will provide a rigorous test of the proposal that hyperfocusing is a key contributor to cognitive dysfunction in schizophrenia and can provide a target for near- and medium-term treatment development efforts.

? DESCRIPTION (provided by applicant): This project continues the development and support of ERPLAB Toolbox, a Matlab-based open source software package for the analysis of event-related potential (ERP) data. ERPs provide information about brain activity related to perception, cognition, emotion, and action with millisecond resolution and are widely used to study a broad range of basic and translational science issues in psychology, neuroscience, psychiatry, neurology, and related fields. There has been an explosion in ERP research, but progress has been hampered by the limited abilities of commercial ERP analysis software and the lack of full-featured open source ERP analysis packages. This situation has slowed the development of new directions in ERP research, instead encouraging researchers to continue with existing approaches supported by the built-in functions of the available commercial analysis packages. ERPLAB Toolbox directly addresses these problems. It adds all the main ERP processing routines to an open source EEG processing package, EEGLAB. ERPLAB builds on EEGLAB by adding powerful tools for the subsequent stages of ERP analysis. All ERPLAB tools can be accessed from the GUI, and they can also be accessed from Matlab scripts to provide automation and customization. ERPLAB provides powerful but easy-to-use tools for the basic and advanced analysis procedures that are commonly used by ERP researchers. In addition, ERPLAB can be easily extended by anyone with rudimentary programming skills, making it possible for researchers to create innovative new data processing and analysis procedures and link ERPs with other types of biological data. ERPLAB has been publicly available for 4 years, and in that time it has been downloaded over 10,000 times and has been cited in 165 publications (a number that is increasing exponentially). It has been used to replace expensive commercial software for conventional analyses, and it has been used as a platform for developing new analysis procedures and for linking ERP data with other open source toolboxes. In the proposed funding period, we will continue to provide basic support and develop new features that reflect trends in basic, translational, and clinical research. We will focus on four themes: increased workflow efficiency; new tools for single-trial analyses; advanced statistical tools; and additional documentation and examples. Our goal is to give researchers free, open-source tools that allow them to conduct innovative, state-of-the-art, high-impact research on normal and disordered brain function.

DESCRIPTION (provided by applicant): Schizophrenia is a prevalent mental health disorder that creates enormous social, economic, and interpersonal hardships for patients and their families. Although hallucinations and delusions are the most salient symptoms of this disease, schizophrenia also involves pervasive cognitive deficits that are key predictors of long-term outcome and are not substantially ameliorated by current medications. Progress in treating these symptoms requires basic science research on the neural and cognitive systems that are dysfunctional in schizophrenia (for the development of targeted treatments) and highly precise measures of these systems (for the assessment of new treatments). The purpose of the present proposal is to advance our understanding of a set of important basic science issues and simultaneously lay the groundwork for the next steps in clinical research. Specifically, the proposed project will explore the mechanisms by which working memory representations control the operation of attention. Current research indicates that this is a key area of dysfunction in schizophrenia, but insufficient basic science is available to guide the next steps of clinical research in this area. In addition, this is a key area of research for understanding the overall architecture of the human mind. The proposed research will explore the processes and circuits by which working memory representations exert control over attention, focusing on the visual modality because of our rich knowledge base about the anatomy, physiology, and function of the visual system. We will use a combination of eye tracking, event-related potentials (ERPs), and psychophysics so that we can precisely determine whether attention is covertly and overtly directed toward items that either match or mismatch items being held in working memory. We will assess both the cognitive processes involved in using working memory to guide attention and the neural circuits that mediate between working memory representations and the implementation of selective attention. This research will feed directly into our program of translational research on cognitive dysfunction in schizophrenia and our efforts to develop next-generation measures that can be used in the development and assessment of new treatments for this disease.

DESCRIPTION (provided by applicant): Attention and emotion have a bidirectional relationship, which serves a strong role in driving human behavior. The purpose of the present proposal is to advance our understanding of the cognitive and neural mechanisms that underlie attention-emotion interactions and individual differences in anxiety. Specifically, the proposed project will use a combination of behavioral and novel event-related potential (ERP) measures to characterize the precise mechanisms and time course of attentional allocation to emotional stimuli. Emotional stimuli that are threatening in nature have been shown to be especially potent modulators of attention. Indeed, abnormal allocation of attention to threatening stimuli has been implicated as a potential etiological factor in anxiety disorders. The proposed research will examine neural and behavioral indices of attention in the context of threat to pinpoint the basic mechanisms that drive attention-emotion interactions and individual differences in anxiety. We will use a well-validated ERP measure of the allocation of attention (the N2pc component) to assess the magnitude and time course of attentional allocation to threatening stimuli. We will also use a newly discovered ERP measure of attentional suppression (the Pd component) to assess the top-down suppression of threatening stimuli. Together, these two ERP measures will make it possible to isolate separate aspects of threat processing that are ordinarily intermixed in behavioral measures. Moreover, the proposed project will examine how these processes vary as a function of individual differences in trait anxiety to characterize the role of atypical attentional processes in characterizing anxiety. This follows from pilot data in which we observed substantial individual differences in ERP indices of suppression of threat, which in turn made it possible to identify individual differences in behavioral measures of threat processing. We will also examine the effects of direct manipulations of state anxiety on these measures. Moreover, because anxiety disorders often involve the acquisition of conditioned emotional responses, we will also examine attention and suppression in the context of conditioned threat. This research will serve an important role in furthering our basic science understanding of the relationships among attention, emotion, and anxiety, and it will also inform future clinical work on anxiety.

? DESCRIPTION (provided by applicant): Schizophrenia is a prevalent mental health disorder that creates enormous social, economic, and interpersonal hardships for patients and their families. Although hallucinations and delusions are the most salient symptoms of this disease, schizophrenia also involves cognitive deficits that predict long-term outcome and are not ameliorated by current medications. In particular, we have shown that schizophrenia patients exhibit large reductions in working memory capacity that predict the degree of overall cognitive dysfunction. Animal models of working memory capacity focus on the idea that working memory is implemented by means of sustained neural activity (active maintenance) that arises from recurrent neural connections, and this has formed the basis of computational models of the microcircuitry of schizophrenia. However, research in humans indicates that both passive and active maintenance mechanisms underlie working memory. These passive and active maintenance mechanisms have different neural substrates and have been hypothesized to play very different roles in broader cognitive function. If we are to understand and treat impaired cognition in schizophrenia, we must develop a better understanding of these active and passive maintenance subsystems. The purpose of the present proposal is to advance our understanding of the active and passive components of working memory in healthy individuals and simultaneously lay the groundwork for the next steps in our program of schizophrenia research. Specifically, the proposed project will test the hypothesis that the active maintenance component of working memory plays the role traditionally ascribed to working memory in general, namely serving as a temporary buffer for non-automated cognitive operations. For example, our preliminary data indicate that the active maintenance of information in a working memory task terminates when a simple discrimination must be performed during the delay period. However, passive maintenance does not appear to be disrupted by the intervening task. Thus, active maintenance is used to perform individual cognitive operations, and passive maintenance is used to maintain other relevant information while these cognitive operations are being performed. Our pilot data also indicate that active maintenance can be used to store precise, metric information about visual features, whereas passive maintenance is more categorical. To differentiate active maintenance from passive storage, we will use multiple converging measures of sustained memory-related brain activity, including sustained electrical potentials in ERP recordings, sustained BOLD activity in fMRI, decoding of memory content from single-trial EEG signals, and decoding of memory content from the single-trial BOLD signal. In addition, advanced psychophysical methods will be used to assess differences in the informational content of active and passive working memory representations. This research will feed directly into our program of translational research on cognitive dysfunction in schizophrenia, providing concepts and methods than can be used in the development and assessment of new treatments.

How do people learn large-scale spaces, like new towns and cities that they visit, as they navigate? Addressing this question poses surprising obstacles, such as the difficulty in optimizing large-scale spaces for experimental testing and controlling for pre-existing knowledge. Desktop virtual reality offers one possible way to address this question, although such testing offers an incomplete rendition of the full-body, immersive experience that is real-world navigation. Researchers will develop a 2-D treadmill coupled with a head-mounted display to allow free ambulation of large-scale virtual spaces. Successful development of this device has important societal applications. For example, pre-training with enriched body-based cues has the potential to increase knowledge transfer to real world environments, which could be helpful for training individuals such as first-responders and navigation in wilderness environments. Also, the device and proposed experiments will provide a completely novel understanding of the neural basis of human spatial navigation with body-based cues, fundamental to accurately modeling spatial cognition and understanding why we often get lost when we visit new cities.

Almost all theories of the neural basis of spatial navigation, largely developed in freely navigating rodents, assume the critical importance of importance of body-based cues to this code. Yet the vast majority of studies in humans involve navigation in desktop virtual reality. The novel device that will be developed will permit 2-D locomotion-based VR navigation, allowing a full range of body/head rotations and ambulation. The experiments will determine 1) the contributions of body-based input to human spatial navigation and how navigation in VR with body-based can enhance subsequent knowledge of real world environments 2) how the brain codes spatial distance by employing simultaneous EEG recordings 3) how the brain codes the relative directions of landmarks in the environment by modeling the underlying multidimensional brain networks using high-resolution functional magnetic imaging (fMRI). The outcomes from these experiments will be important to testing models of spatial navigation and advancing our understanding of the extent to which we employ visual vs. body-based cues to represent spatial environments, currently an issue of significant debate in the field.

? DESCRIPTION (provided by applicant): Schizophrenia is a prevalent mental health disorder that creates enormous social, economic, and interpersonal hardships for patients and their families. Although hallucinations and delusions are the most salient symptoms of this disease, schizophrenia also involves cognitive deficits that predict long-term outcome and are not ameliorated by current medications. In particular, we have shown that schizophrenia patients exhibit large reductions in working memory capacity that predict the degree of overall cognitive dysfunction. Animal models of working memory capacity focus on the idea that working memory is implemented by means of sustained neural activity (active maintenance) that arises from recurrent neural connections, and this has formed the basis of computational models of the microcircuitry of schizophrenia. However, research in humans indicates that both passive and active maintenance mechanisms underlie working memory. These passive and active maintenance mechanisms have different neural substrates and have been hypothesized to play very different roles in broader cognitive function. If we are to understand and treat impaired cognition in schizophrenia, we must develop a better understanding of these active and passive maintenance subsystems. The purpose of the present proposal is to advance our understanding of the active and passive components of working memory in healthy individuals and simultaneously lay the groundwork for the next steps in our program of schizophrenia research. Specifically, the proposed project will test the hypothesis that the active maintenance component of working memory plays the role traditionally ascribed to working memory in general, namely serving as a temporary buffer for non-automated cognitive operations. For example, our preliminary data indicate that the active maintenance of information in a working memory task terminates when a simple discrimination must be performed during the delay period. However, passive maintenance does not appear to be disrupted by the intervening task. Thus, active maintenance is used to perform individual cognitive operations, and passive maintenance is used to maintain other relevant information while these cognitive operations are being performed. Our pilot data also indicate that active maintenance can be used to store precise, metric information about visual features, whereas passive maintenance is more categorical. To differentiate active maintenance from passive storage, we will use multiple converging measures of sustained memory-related brain activity, including sustained electrical potentials in ERP recordings, sustained BOLD activity in fMRI, decoding of memory content from single-trial EEG signals, and decoding of memory content from the single-trial BOLD signal. In addition, advanced psychophysical methods will be used to assess differences in the informational content of active and passive working memory representations. This research will feed directly into our program of translational research on cognitive dysfunction in schizophrenia, providing concepts and methods than can be used in the development and assessment of new treatments.