Antonio Damasio, MD PhD - US grants

During the current funding period, we have formulated a theory to account for the defects in decision-making, planning and social conduct which are so salient in patients with frontal lobe lesions. It suggests that those defects result from defective activation of somatic markers that normally function as signposts for making decisions advantageous for the organism. The theory, based on previous data and on principles outlined in the SMR framework discussed in Project #1, will form the basis for testing a series of specific hypotheses regarding the composition of the neural network presumed to underlie the process of somatic marker activation. Psychophysiological measures (skin conductance, heart rate, facial EMG) will be used as indexes of somatic state activation, and we will study patients with focal lesions in the regions hypothesized to comprise the somatic state neural network. Among them are the ventromedial frontal, dorsolateral prefrontal, and somatosensory cortices. We will also investigate the judgment and decision-making capacities of target patients in a set of field experiments, and we will study the psychopathological manifestations of such patients using a comprehensive set of standard inventories and rating scales. Elucidation of the neural substrates of "executive functions" will advance neuroscience by contributing fundamental knowledge regarding decision- making and planning in social and other realms, and by promoting better diagnosis of the numerous patients who suffer from disturbances of executive function caused by neurological diseases such as head injury, stroke, and surgical ablation of tumors. Knowledge gained from this project will also permit the design of rehabilitation techniques which may allow patients to compensate for their defects and obtain gainful employment and more rewarding interpersonal relationships. Other applications include the extension of the theoretical constructs of this project to conditions such as psychopathy, obsessive-compulsive disorder, and the psychosocial consequences of chronic alcoholism and other forms of addiction. We anticipate that the results will have an impact on the field of education, and on the management of individuals with developmental forms of social conduct defect.

Drawing on a theory of the neural architecture and operation underlying memory developed during the current funding period, we plan to study patients with a broad range of disorders of learning and memory, especially those with memory impairments circumscribed to a sensory modality or to a domain of knowledge. The goal is to parcellate the neural substrates of what is traditionally known as semantic (generic) and episodic memory, by studying neural systems underlying recall and recognition for varied knowledge domains and for different levels of taxonomic complexity. We aim at testing a number of hypotheses derived from our theoretical framework, regarding the recall and recognition impairments that are likely to be observed in patients with damage to various neural fields in medial temporal, anterolateral temporal, and occipitotemporal regions. In turn, the evidence uncovered in these studies will help optimize the evolving theoretical account. In addition to furthering understanding of the neural substrates of learning and memory, which is a core theme of neuroscience, elucidating the neural basis of memory has several practical consequences. It will promote better diagnosis of the numerous patients who suffer from memory impairments caused by neurological disease, including head injury, stroke, and Alzheimer's disease, and permit the design of rehabilitation techniques which may partially offset their defects. Knowledge gained from our studies may enhance the design of tasks in which humans are required to learn and recognize, and optimize the design of computational systems capable of learning and recognizing information. Our approach is aimed at both the structural and psychological levels (i.e., neuroanatomical systems and cognitive processes), we are in an advantageous position to capitalize on the unique opportunities affordable by neuropsychological experimentation in humans with brain damage.

The Core project provides a central resource for the Program. It encompasses four Subcore Units, which perform the following specific operations: (1) Image Acquisition and Analysis; (2) Statistical Analysis, Data Storage, and Computer Facilities; (3) Subject Recruitment and Induction; and (4) Monitoring of Data Quality and Research Progress. All of these operations, which are essential for the execution of all studies involving humans, are continuous over time. This team of investigators has an established record of carrying out such activities in other research protocols, for more than a decade.

Preliminary evidence from neurological patients with lesions and from normal individuals studied with PET, suggests that the normal process of retrieving words which denote concrete entities depends on multiple regions of the left cerebral hemisphere located outside the classic language areas. Moreover, at the level of large-scale systems, there appears to be an intriguing and principled relationship between the anatomical region and the kind of item being named. Drawing on these findings, we have developed a set of hypotheses regarding the neural basis of the process of lexical retrieval. In this Program, we propose to investigate these hypotheses in both English and American Sign Language (ASL), using the same approach - functional imaging - and the same paradigm - [15-O] H2O - PET activation studies. The Program brings together investigators from two institutions, diverse fields of expertise, and diverse human populations of normal hearing, deaf, and neurological patients. Results from the investigations proposed here promise to broaden the currently limited knowledge available on the neurobiology of language. Such new knowledge is critical to further the understanding of communication disorders in both the normally hearing and the deaf, and is indispensable for the development of effective rehabilitation strategies for aphasics in both auditory-based and sign languages. The knowledge gathered here can also be used to constrain ongoing theoretical and empirical work on the linguistic and cognitive aspects of the lexicon.

The Core project is the central resource of the Program. It encompasses five Core Units, which carry out the following operations: (a) Patient Registry, which regards the identification, screening and induction of subjects into the Registry, and the central Database and data management for neuroanatomical and neuropsychological data; (b) performance of neuroimaging studies and neuroanatomical analysis at both the screening/induction and chronic stages; (c) performance of basic neuropsychological studies at the time of screening/induction and at the chronic stage; (b) statistical analysis as needed by different projects; (c) monitoring of all matters pertaining to subject confidentiality, informed consent procedures, data quality and research progress. These operations, which are essential for the execution of all studies in our Program involving humans, are carried on continuously. They have been carried out smoothly, in all aspects, for the duration of the Program, and remain unchanged for the forthcoming period. The investigators in the Program have also continued to develop new techniques required to improve the quality of neuroanatomical and neuropsychological data gathering and analysis. As a result, new products are now available, to this research group and to others, for cognitive neuroscience research (e.g., single-channel tissue segmentation, 3- dimensional lesion analysis in a standard anatomical space (the MAP-3 technique)).

[unreadable] DESCRIPTION (provided by applicant): [unreadable] This is a request for renewal of a Program on the neuroanatomical basis of cognition and behavior initiated in 1983, and renewed in 1986, 1991, 1996, and 2001. From the outset, the purposes of the Program have been: (1) to contribute to the understanding, diagnosis, and management of brain disease; (2) to elucidate, in hypothesis-driven research, the relation between large-scale human neural systems and functions such as emotion, decision-making, memory, and language; and (3) to maintain and expand a neurological Patient Registry, which constitutes a unique resource for cognitive neuroscience. The research relies on neurological lesions used as probes to test hypotheses regarding the role of cortical and subcortical systems involved in complex functions. The lesions are defined by in vivo neuroimaging techniques, and the functional impairments are characterized by neuropsychological and psychophysiological techniques. The approach is now strengthened by studies aimed at elucidating the pattern of interconnecting pathways in neural systems, and by the development of powerful new imaging and lesion analysis tools. The investigators have obtained results that have influenced cognitive neuroscience, and have provided a body of knowledge applicable to diagnostic procedures and valuable to inform the design of rehabilitation approaches. The renewal of this Program is likely to yield results of scientific merit and practical importance, potentially even stronger than in the past. The new proposal identifies exciting new questions to pursue, in lesion analysis, neuroeconomics, emotional and social development, social emotions, and memory and language disorders; it demonstrates the stability and cohesiveness of the senior research team, combined with the addition of a performance environment which will strengthen the intellectual and technical resources; and it maintains and expands the highly valuable resource of the Patient Registry. These factors join to provide ideal and compelling circumstances for continuation of the Program. [unreadable] [unreadable] Public health relevance: This research will provide information critical for the diagnosis and treatment of neurological patients who suffer stroke, head injury, brain tumors, seizure disorders, and other brain diseases. The information will help with the design of rehabilitation programs that can alleviate deficits in memory, language, and emotions, and help curb the suffering associated with brain disease. [unreadable]

A multidisciplinary research team will study various aspects of the nature of moral judgments and the causal factors for the capacity for cross-cultural variation and change. The project measures the nature of moral decisions across different time periods (evolutionary, developmental, and cultural) and among different test populations (nonhuman animals, normal and neurobiologically impaired human infants and adults, and different cultures). It uses different methods for each type of study including experiments of primates, large-scale internet studies, and neuropsychological investigation of patients. The investigators will to study two psychological factors: (1) The idea that intending to harm another as a means to the greater good is less permissible than harming merely as a foreseen side effect (the intention principle) and (2) The idea that acts that cause harm to others will be perceived as morally worse than omissions of an act that causes equivalent harm (the omission principle). Studies of these principles will be conducted with nonhuman primates and human infants to test the hypothesis that some of the core cognitive building blocks that are necessary for these principles (e.g., perceiving intentions and goals) are in place but only take on moral significance in our own species, and only later in child development. The investigators will test the hypothesis that these principles are universal but with cross-cultural variation in their specific content (e.g., who can be harmed) by using both large-scale internet-based studies as well as studies of hunter-gather and subsistence-based societies. They will test the hypothesis that governments can impose explicit laws that alter how people behave yet these explicit norms do not penetrate people's intuitive moral judgments. The investigators also will examine how neural insult systematically changes the nature of particular moral judgments among patient populations (i.e., autistics, individuals with damage to the frontal lobes and amygdala).

The project is expected to enhance basic understanding of how humans evolved the capacity to deliver moral judgments, how such judgments change over development and across cultures, and how the capacity breaks down following selective neural insult. Results from this project are likely to be useful in the arenas of justice, public policy, education, and clinical treatment, showcasing the biological and psychological mechanisms that humans bring to the moral table, and how they respond to policy that may be at odds with their intuitive moral sense. The project also will provide education and training opportunities for graduate students, including students from minority groups and developing nations.

With support from the Robust Intelligence program in the Division of Information and Intelligent Systems and the NSF 2026 Fund Program in the Office of Integrated Activities, investigators at University of Southern California are investigating a radical new approach to creating machines with the foundations for nature-inspired feeling as a primary basis for its action in the world. Recent theoretical work has raised the possibility that an analogue of feelings could be constructed in robots with bodies that are physically vulnerable to the environment. The possession of a body that can exist in better or worse condition, and the capacity to recognize and respond to such states are key elements in the generation of feeling in living creatures. This project will study a new class of machines organized according to the principles of life regulation, or homeostasis. The fundamental innovation of these machines is the introduction of risk-to-self. This project will explore how robots might respond to risk-to-self based on homeostasis rather than rely on the robustness provided primarily by e.g. increases in computation or protective encasement. The work aims to explore the nature-inspired question of the material basis for feelings and their potential value to an AI agent. By investigating machine-based homeostasis as the foundation of machine ?feeling? this project might lay the foundations for safer and perhaps even ?empathetic? robots in the spirit of the 2026 Idea Machine entry, ?Promoting Empathy-Based AI?.

This project will integrate recent breakthroughs from two disparate research fields: soft robotics and deep learning for multisensory integration. The field of soft robotics presents exciting new opportunities for machines to generate homeostatic data that are far richer than those achievable in rigid-bodied robots. The special properties of soft materials allow for greater flexibility, compliance, self-repair, and dense multimodal sensing. The tools of deep learning will be specifically applied to build multimodal representations that can bridge across internal and external sensory representations. The studies will be performed in computer simulations of voxel-based soft robots controlled by cross-modal neural networks. The behavioral advantages that accrue to soft robots possessing a homeostatic architecture will be quantified, both for robots in isolation and for societies of robots. This project's simulation work will lay the necessary preparations for future studies of machine feeling in physically realized robots.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.