Bayne, T., Frohlich, J., Cusack, R., Moser, J., & Naci, L. (2023). Consciousness in the cradle: on the emergence of infant experience. Trends in Cognitive Sciences.
Bayne, Tim, et al. “Consciousness in the cradle: on the emergence of infant experience.” Trends in Cognitive Sciences (2023).
@article{bayne2023consciousness, title={Consciousness in the cradle: on the emergence of infant experience}, author={Bayne, Tim and Frohlich, Joel and Cusack, Rhodri and Moser, Julia and Naci, Lorina}, journal={Trends in Cognitive Sciences}, year={2023}, publisher={Elsevier} }
New research sheds fresh light on mystery of infant consciousness
When does consciousness begin? There is evidence that some form of conscious experience is present by birth, and perhaps even in late pregnancy, an international team of researchers led by Tim Bayne of Monash University in Melbourne, Australia and Joel Frohlich of the University of Tuebingen in Germany and the US-based Institute for Advanced Consciousness Studies in Santa Monica, California has concluded in a new review manuscript. The findings, just published in the peer-reviewed journal ‘Trends in Cognitive Science’, have important clinical, ethical, and potentially legal implications, according to the authors.
In the study, entitled ‘Consciousness in the cradle: on the emergence of infant experience’, the researchers argue that, by birth, the infant’s developing brain is likely capable of conscious experiences. Although each of us was once a baby, infant consciousness remains mysterious, because infants cannot tell us what they think or feel, explains one of the two lead authors of the paper Dr. Tim Bayne, Professor of Philosophy at Monash University.
“Nearly everyone who has held a newborn infant has wondered what, if anything, it is like to be a baby. But of course we cannot remember our infancy, and consciousness researchers have disagreed on whether consciousness arises ‘early’ (at birth or shortly after) or ‘late’ – by one year of age, or even much later.”
To provide a new perspective on when consciousness first emerges, the team reviewed recent advances in consciousness science. In adults, some markers from brain imaging have been found to reliably differentiate consciousness from its absence, and are increasingly applied in science and medicine. This is the first time that these advances, as translated to infants, have been reviewed in detail.
Co-author of the study, Dr. Lorina Naci, Associate Professor at Trinity College Dublin in Ireland, who leads the ‘Consciousness and Cognition Group’, explained: “Our findings suggest that newborns can integrate sensory and developing cognitive responses into coherent conscious experiences to understand the actions of others and plan their own responses.”
It is even possible that birth itself triggers the onset of consciousness. “Probably the first thing the newborn infant realizes is that the outside world is very unpredictable relative to the womb environment,” explained co-lead author and postdoctoral researcher Dr. Joel Frohlich. “Things are constantly changing, and so the newborn must build a mental model of the world to adapt and predict things.”
However, the authors don’t rule out the possibility that consciousness might already start some weeks beforehand. “Julia Moser’s work shows that third-trimester fetuses appear to be capable of learning sequences of auditory beeps,” said Dr. Frohlich, referring to his co-author Dr. Moser at the University of Minnesota. “When an auditory tone deviates from a pattern established earlier in the experiment, the fetus shows this ‘surprise’ response in its magnetic brain activity. The neural activity shows a field deflection as if the fetus is saying ‘huh?’.”
The paper also sheds light into ‘what it is like’ to be a baby. We know that seeing is much more immature in babies than hearing, for example (see the image below for a theoretical depiction). Furthermore, this work suggests that, at any point in time, infants are aware of fewer items than adults, and can take longer to grasp what’s in front of them, but they can easily process more diverse information, such as sounds from other languages, than their older selves.
“Infants can perceive many things which adults cannot, like the differences between vowel sounds in a foreign language,” explained Dr. Joel Frohlich. “By 10 months or so, we lose this ability as the brain decides these perceptual differences are no longer relevant and discards them.”
This blog post is based on a recent book chapter “VR for Cognition and Memory” in Current Topics in Behavioral Neuroscience: Virtual Reality in Behavioral Neuroscience: New Insights and Methods. This work presents a review of research on VR’s ability to provide ecologically valid environments to study memory and cognition and discusses how features like interactivity, locomotion, and contextual control engage the brain’s memory systems more naturally than lab studies.
Reggente N. (2023). VR for Cognition and Memory. Current topics in behavioral neurosciences, 10.1007/7854_2023_425. Advance online publication. https://doi.org/10.1007/7854_2023_425
Reggente N. VR for Cognition and Memory [published online ahead of print, 2023 Jul 14]. Curr Top Behav Neurosci. 2023;10.1007/7854_2023_425. doi:10.1007/7854_2023_425
Reggente, Nicco. “VR for Cognition and Memory.” Current topics in behavioral neurosciences, 10.1007/7854_2023_425. 14 Jul. 2023, doi:10.1007/7854_2023_425
Revolutionizing Cognition Research with Virtual Reality
For decades, scientists have worked tirelessly to elucidate the intricate neural machinery supporting human cognition. This endeavor is certainly not for the faint of heart, as formidable challenges present themselves at every turn.
“To study cognition holistically means investigating interconnections between its rich repertoire of functions, including attention, reasoning, language, and memory. Memory is a particularly crucial facet, as it supports and subserves all other aspects of cognition; no cognitive task can be accomplished without memory.”
A holistic understanding demands that we study cognition as it operates in its natural habitat – the real world. Otherwise, as the parable of the blind men and the elephant warns, we risk gross mischaracterizations. Researchers must therefore conduct experiments in “verisimilar contexts (i.e. contexts appearing as the RW)” to achieve ecological validity.
Virtual reality (VR) presents an unprecedented opportunity in this regard. By simulating the real world, we can now study memory and cognition with enhanced veridicality.
“The environmental customization afforded by VR makes it an ideal tool for studying cognition in an ecologically valid fashion. Through the lens of memory studies, this chapter showcases the ways in which VR has advanced a meaningful and applicable understanding of cognition.”
The article presents a thorough review of research that showcases how VR is revolutionizing the study of cognition and memory.
Bridging the Gap Between Lab and Real-World Cognition
Traditional lab experiments often possess limited generalizability, whereas VR can provide naturalistic environments and tasks that echo real-world demands, easily bolstering ecological validity. Previous work has made a compelling case for how VR enhances the ecological validity of fMRI memory research.
VR experiences engage recollection-based memory retrieval akin to real events, unlike lab stimuli which rely more on familiarity. Indeed, VR experiences appear to be retrieved via recollection-based processes similar to those that support autobiographical/recollection memory, whereas retrieval of conventional screen experiences seems more similar to familiarity. This makes VR apt for integrated cognition and memory research.
VR Permits for Information to be Situation in Space
Most importantly, VR permits realistic navigation around virtual environments (c.f.), affording users with a sense of space (the scaffolding of memory). Both philosophers and psychologists alike postulate that brains have evolved solely to support purposeful and predictable movement. Many posit that the ontogeny of episodic memory relates to the onset of locomotion during infancy that scales with Hippocampal development (which also provides a mechanism for infantile amnesia and age-related episodic memory loss). One source of evidence to support this proposition is in the life cycle of the bluebell tunicate. This filter feeder begins to digest a substantial chunk of its cerebral ganglion once identifying a suitable undersea perch to spend the rest of its existence. This phenomenon suggests that once it has served its purpose as a neural network supporting movement, the cerebral ganglion yields greater utility to the organism as nutrition.
From chemotaxis to cognitive maps, a representation of space is necessary for meaningful movement. A neural instantiation of a map that provides spatial bookmarks of an organism’s experiences, demarcating the locations of nutrition and enemies within an environment, is a fundamental component of brains. Indeed, there is a primacy of spatial content in the neural representation of events. Spatial information is often recalled earliest in the retrieval process, and the degree to which individuals report confidence in their autobiographical memories is predicted by their knowledge of the spatial layout of the setting in which the memory occurred. The Method of Loci (a.k.a. Memory Palace) mnemonic has long been appreciated for its ability to increase memory by imagining to-be-remembered information placed at familiar locations. Past work used a VR implantation of this technique to suggest that the principal component behind mnemonic efficacy is the explicit binding of the objects to a spatial location and revealed a tight relationship between spatial memory (SM) and free recall of encoded objects. These observations showcase that space and memory are inextricably linked at conceptual and neuronal levels – a notion that has become entrenched in popular culture; the phrase “out of space” is often used when indicating a computer’s memory is full.
If space is the inescapable wallpaper that serves as the backdrop for all experience, then it follows that as our spatial or environmental context changes, so should the neural activity underlying diverse cognitive processes. Given that VR can easily change environments, it provides an unparalleled landscape with which to study the intersection of space, memory, and cognition.
Additionally, VR enables human analogs of spatial memory research previously limited to animal models, like virtual radial arm mazes. This facilitates powerful translational research from rodents to humans.
Key Features of VR That Facilitate Cognition Research
Below are some features highlighted by the chapter that are exclusive to VR. Such features permit real-world scenarios with increased experimental control and significantly less costs.
VR provides absolute control over the environment. This permits isolation and systematic manipulation of spatial contexts, immersion, emotions, embodiment, etc.
Rapid teleportation between environments induces robust context-dependent learning, a fundamental principle in memory encoding.
Interactivity and locomotion increase embodiment and navigational involvement, enhancing hippocampal memory systems.
Implicit metrics like gaze, paths, and object interactions generate objective measures of memory and attention unbiased by subjective reporting.
Brain imaging during VR reveals in vivo neural correlates of cognition impossible with real-world navigation.
VR spatial mnemonics such as the Method of Loci can provide performance improvements over just imagination by standardizing and controlling the environments.
Applications of VR for Assessing and Enhancing Cognition
Conventional measures of memory typically focus on core content (i.e., the “what”) instead of the true binding that happens in actual episodes (i.e., “what,” where,” and “when”). They also often use verbal materials, which makes the test sensitive to performance in non-memory domains, permitting for compensatory strategies which could erroneously reveal normal “memory.” Subjective reports rarely scale with performance on traditional memory tests, warranting criticism that such measures wrongly estimate memory capacities for everyday situations. For example, patients reporting topographical memory deficits have preserved ability in tabletop tests of spatial or geographical knowledge. Additionally, cognitive complaints in amnesiacs typically show little correlation with verbal memory tests used in clinical settings.
VR tasks, however, have been more reliable in tracking self and caregiver reports of deficits that impact quality of life. The points below highlight other aspects of VR that can increase the ecological validity of both the detection and amelioration of memory deficits.
VR scenarios like virtual stores and routes enable sensitive, ecologically valid tools to identify mild cognitive impairment early.
VR spatial navigation paradigms can differentiate Alzheimer’s from milder impairment based on hippocampal recruitment patterns.
VR enables safe exposure therapy for memory deficits induced by trauma and realistic training for brain injury rehabilitation.
Spatial mnemonic techniques adapted to VR boost memory beyond baseline abilities in healthy individuals.
VR puzzles engage aging minds, increasing motivation. Long-term regimes may prevent decline. As one study found, “6 months of VR training powerfully increased long-term recall.”
VR training could augment real-world cognition and rehabilitate deficits, with proven memory transfer effects.
In conclusion, VR enables an unprecedented ability to understand real-world cognition, precisely diagnose impairments, and develop interventions that enhance memory and cognition. The immersive, interactive nature of VR environments engages our brains’ memory systems far more naturally than traditional lab studies.
The inherently engaging qualities of VR, coupled with its ability to implicitly quantify and enhance memory, make it a powerful tool in populations spanning from pediatrics to the elderly.
Indeed, VR may catalyze discoveries about the very mechanisms underlying human consciousness itself, which intimately relies on episodic memory. By augmenting these processes, VR could profoundly transform our experience and understanding of consciousness. The future of cognition research has never looked more exciting.
Preventing antisocial robots: A pathway to artificial empathy at Science Robotics
Cite This Work
APA
MLA
Chicago
Harvard
Vancouver
Christov-Moore, L., Reggente, N., Vaccaro, A., Schoeller, F., Pluimer, B., Douglas, P. K., Iacoboni, M., Man, K., Damasio, A., & Kaplan, J. T. (2023). Preventing antisocial robots: A pathway to artificial empathy. Sci. Robot, 8, eabq3658. https://doi.org/10.1126/scirobotics.abq3658
Christov-Moore, Leonardo, et al. “Preventing Antisocial Robots: A Pathway to Artificial Empathy.” Sci. Robot, vol. 8, eabq3658, 2023, https://doi.org/10.1126/scirobotics.abq3658.
Christov-Moore, Leonardo, Nicco Reggente, Anthony Vaccaro, Felix Schoeller, Brock Pluimer, Pamela K. Douglas, Marco Iacoboni, Kingson Man, Antonio Damasio, and Jonas T. Kaplan. “Preventing Antisocial Robots: A Pathway to Artificial Empathy.” Sci. Robot 8 (2023): eabq3658. https://doi.org/10.1126/scirobotics.abq3658.
Christov-Moore, L., Reggente, N., Vaccaro, A., Schoeller, F., Pluimer, B., Douglas, P. K., Iacoboni, M., Man, K., Damasio, A., & Kaplan, J. T. (2023). Preventing antisocial robots: A pathway to artificial empathy. Sci. Robot, 8, eabq3658. https://doi.org/10.1126/scirobotics.abq3658
Christov-Moore L, Reggente N, Vaccaro A, Schoeller F, Pluimer B, Douglas PK, Iacoboni M, Man K, Damasio A, Kaplan JT. Preventing antisocial robots: A pathway to artificial empathy. Sci. Robot. 2023;8:eabq3658. doi:10.1126/scirobotics.abq3658.
Look, whether you’re a doomer or a techno-utopian, whether you were ready or not, the age of artificial intelligence (AI) probably arrived sometime in this decade. This age brings deep, important, and melancholy reflections on intelligence, creativity, and what it is to be human. However, If we can’t ensure that AI is aligned with human interests, we may have little time to reflect. Containment, or a giant pause button, is not a likely option. There is too much real-world inertia and distrust among world actors to ensure everyone will comply – and it only takes one successful experiment to unleash a truly unforeseen problem into the world. In a new paper in Science Robotics, we tackle this problem through three big ideas, that we’ll call the problem, the path, and the potential.
The Problem
There is a pressing need to imbue AI with a value system that allows it to “understand” harm in way that inherently demotivates it from making catastrophic, irreversible decisions, without the need for complex rule systems. This value system must scale with AI’s rapid self-improvement and adaptations as it encounters novel situations and greater responsibilities for peoples’ well-being. Biology suggests that empathy could provide this value. Empathy allows us to understand and share the feelings of others, motivating us to alleviate suffering and bring happiness.
However, most approaches to artificial empathy focus on allowing AI to decode internal states and act empathetically, neglecting the crucial capacity for shared feeling that drives organisms to care for others. Here lies the problem: Our attempt to create empathic AI may inadvertently result in agents that can read us perfectly and manipulate our feelings, without any genuine interest in our wellbeing, or understanding of our suffering. Our well-meaning attempts to produce empathy may produce superintelligent sociopaths.
The Path Towards Artificial Empathy
If we are giving birth to the next form of life, it’s not far-fetched to see ourselves as collective parents, with a civilizational responsibility. When you’re raising something as potentially powerful as AI, what should you do? The formative years of powerful yet ethical figures like Buddha, Jesus (or Spiderman) teach us that the responsibility of great power is learned by experiencing the suffering that all living beings endure. Power without vulnerability and compassion can easily cause harm, not necessarily through malice, but through obliviousness or an unconstrained drive to fulfill desires.
To address this, we propose a speculative set of guidelines for future research in artificial empathy. Firstly, even if it’s only during a specific phase of their training, AI need to possess a vulnerable body that can experience harm, and learn to exist in an environment where actions have consequences for its physical integrity. Secondly, AI should learn by observing other agents and understanding the relationship between their experiences and the state of their own bodies, similar to how it understands itself. Lastly, AI should learn to interact with other agents in a way that avoids harm to itself and others. Perhaps it will emergently behave in a more ethical fashion if harm to others is processed like harm to itself. Vulnerability is the common ground from which genuine concern and aversion to harm naturally emerge.
The Potential of Artificial Empathy
Achieving true artificial empathy could transform AI from a potential global threat to a world-saving ally. While human empathy is crucial in preventing harm and promoting prosocial behavior, it is inherently biased. We tend to prioritize the suffering of a single relatable person over the plight of a stranger or very large numbers of people. This bias arises due to our brain’s difficulties in handling the large-scale, long-term, and nonlinear problems often encountered in complex societies. The scalable cognitive complexity of an empathic AI might be capable of proposing compassionate solutions to these grand challenges that surpass the human capacity for comprehension and imagination. However, every solution brings new challenges. How can we trust an intelligence that surpasses our own? What sort of responsibilities will we have for an intelligence that can suffer?
If we are the collective parents to a new superbeing, we must decide, right now, what kind of parents we are going to be, and what kind of relationship we want with our progeny. Do we want to try and control something we fear, or do the work to raise someone we can trust, to care for us in old age?
If we are the collective parents to a new superbeing, we must decide, right now, what kind of parents we are going to be, and what kind of relationship we want with our progeny. Do we want to try and control something we fear, or do the work to raise someone we can trust, to care for us in old age? Let’s be far-fetched for a short moment: maybe we can guide the development of the upcoming superintelligences toward what Buddhist scholars call “metta,” a cultivation of universal compassion for all beings. Maybe the next Buddha will be artificial.
We are grateful to the Templeton World Charity Foundation and Tiny Blue Dot Foundation for making this work possible. We also extend our thanks to the Survival and Flourishing Fund for their recent award, which will enable us to implement these ideas in simulations with the assistance of talented researchers such as Adam Safron, Guillaume Dumas, and Zahra Sheikh. You can keep track of our latest developments on our artificial empathy project page.
