Periodic Reporting for period 2 - AVIAN MIND (Inquiries Into a Different Kind of Mind)
Reporting period: 2023-07-01 to 2024-12-31
• What is the problem/issue being addressed?
Mammals inhabit virtually all ecological niches in which vertebrates can survive, and wherever they occur, they are the most important predators. This evolutionary success is due in part to the large size of many mammalian brains that enables higher cognitive performance. In other words, an increase in the number of cortical neurons allows greater cognitive processing capacity. A second fundament of the cognitive prowess of mammals are the anatomical intricacies and resulting processing dynamics of the mammalian cortex, which guarantees exceptional cognitive abilities. With its laminar and columnar architecture, the neurons of the cortex process incoming information along the radial dimension and associate it tangentially with distant brain areas.
By studying brain and cognition in diverse mammalian species, we have now learned so many details about the relationship between cortical structure, brain size and function that it is hard to imagine that the identical cognitive functions can be produced with brains that are up to 40x smaller and have no cortex at all. But this is exactly the case. Birds are cognitively on a par with mammals, and some bird taxa even reach the cognitive level of great apes. And that with much smaller brains and a seemingly unstructured nuclear pallium. My project aims to understand how this is possible.
I am convinced that by scientifically explaining how complex cognition can arise from small, non-cortical brains, we will gain a deeper understanding of the crucial mechanisms that define the neuronal fundaments of cognition. This approach will also help to identify which neuronal properties are biological “conditio sine qua non”-components of complex cognition. We can identify them since they should occur in highly similar ways both in mammals and in birds. In contrast, for other cognitive functions birds and mammals might have developed different cellular mechanisms. The first components would then be “hard-to-replace neural systems for cognition”, which have developed convergently during evolution in both mammals and birds. The second group of neuronal circuits, on the other hand, would provide cognitive services that can be realized with a wide variety of properties.
• Why is it important for society?
We humans created the Anthropocene (for better or for worse) by the unmatched power of our cognition. It is this ability that makes us human. The more astonishing is it to realize that studies on avian cognitive neuroscience challenge core assumptions of our understanding how complex cognition is realized in brains. I propose that a comparative approach to study the neural fundaments of cognition in two highly cognitive taxa (mammals and birds) that are separated by 324 million years of brain evolution will provide key insights into common core mechanisms of cognition. This course of research enables an identification of ‘hard-to-replace’ neural designs for cognition that might even also be discovered outside the vertebrate realm like in bees or octopods. Thus, to more fundamentally understand the link between brain structure and cognitive function, we have to leave the trodden path of always analysing the same small number of mammalian species. This is what I’m doing
The project conceptualizes avian executive functions as an ongoing interaction within a network of functionally specialized nodes. Thus, it shifts from a focus on single areas to an emphasis of networks and takes the field of comparative functional neuroscience to the next conceptual level. To this end, it introduces with awake bird fMRI a novel and ground-breaking platform that will importantly expand experimental possibilities in the field.
By questioning the assumption that the avian hippocampus is related to overall formation of new memories, the project opens the possibility and takes the first steps towards a novel vertebrate memory system that is not bound to the hippocampus. This has the potential to start a completely new avenue of research on the neural fundaments of memory.
If small brained, non-cortical pigeons would fulfil the same behavioural, electrophysiological, and imaging-based criteria that are typical for conscious experiences in humans and monkeys, we might be inclined to conceive consciousness as a mental state owned by many animals. This does not imply that pigeon consciousness “feels the same” as in humans. But it would mean that some conscious (for lack of a better word) self-experience of an organism, however simple, would be given to many, if not most animals. Such a data set would have an impact that goes much beyond cognitive neuroscience and should stir discussions from the humanities up to fields like animal ethics.
• What are the overall objectives?
In my project, I aim to implement three conceptual changes:
- To understand the neural basis of avian executive functions, we need to move from studying individual areas to cognition-specific pallial networks that together form the executive system. The activation of this system could then be accompanied by the deactivation of a bird-specific default mode network. To test these hypotheses, I will identify and analyze these network structures and their components using cognitive experiments, fMRI, electrophysiology and optogenetics.
- Key components of the avian hippocampus differ from those of mammals, so that avian memory may be less dependent on the hippocampus. I will use fMRI-based memory tasks and optogenetics to test whether birds have indeed evolved a completely different memory system.
- Consciousness, albeit in simpler forms, could be a mental state that many animals, including birds, possess. Here I will test whether neural correlates of consciousness, as shown in humans and monkeys, can also be demonstrated in pigeons using behavioral studies, fMRI and electrophysiology.
Mammals inhabit virtually all ecological niches in which vertebrates can survive, and wherever they occur, they are the most important predators. This evolutionary success is due in part to the large size of many mammalian brains that enables higher cognitive performance. In other words, an increase in the number of cortical neurons allows greater cognitive processing capacity. A second fundament of the cognitive prowess of mammals are the anatomical intricacies and resulting processing dynamics of the mammalian cortex, which guarantees exceptional cognitive abilities. With its laminar and columnar architecture, the neurons of the cortex process incoming information along the radial dimension and associate it tangentially with distant brain areas.
By studying brain and cognition in diverse mammalian species, we have now learned so many details about the relationship between cortical structure, brain size and function that it is hard to imagine that the identical cognitive functions can be produced with brains that are up to 40x smaller and have no cortex at all. But this is exactly the case. Birds are cognitively on a par with mammals, and some bird taxa even reach the cognitive level of great apes. And that with much smaller brains and a seemingly unstructured nuclear pallium. My project aims to understand how this is possible.
I am convinced that by scientifically explaining how complex cognition can arise from small, non-cortical brains, we will gain a deeper understanding of the crucial mechanisms that define the neuronal fundaments of cognition. This approach will also help to identify which neuronal properties are biological “conditio sine qua non”-components of complex cognition. We can identify them since they should occur in highly similar ways both in mammals and in birds. In contrast, for other cognitive functions birds and mammals might have developed different cellular mechanisms. The first components would then be “hard-to-replace neural systems for cognition”, which have developed convergently during evolution in both mammals and birds. The second group of neuronal circuits, on the other hand, would provide cognitive services that can be realized with a wide variety of properties.
• Why is it important for society?
We humans created the Anthropocene (for better or for worse) by the unmatched power of our cognition. It is this ability that makes us human. The more astonishing is it to realize that studies on avian cognitive neuroscience challenge core assumptions of our understanding how complex cognition is realized in brains. I propose that a comparative approach to study the neural fundaments of cognition in two highly cognitive taxa (mammals and birds) that are separated by 324 million years of brain evolution will provide key insights into common core mechanisms of cognition. This course of research enables an identification of ‘hard-to-replace’ neural designs for cognition that might even also be discovered outside the vertebrate realm like in bees or octopods. Thus, to more fundamentally understand the link between brain structure and cognitive function, we have to leave the trodden path of always analysing the same small number of mammalian species. This is what I’m doing
The project conceptualizes avian executive functions as an ongoing interaction within a network of functionally specialized nodes. Thus, it shifts from a focus on single areas to an emphasis of networks and takes the field of comparative functional neuroscience to the next conceptual level. To this end, it introduces with awake bird fMRI a novel and ground-breaking platform that will importantly expand experimental possibilities in the field.
By questioning the assumption that the avian hippocampus is related to overall formation of new memories, the project opens the possibility and takes the first steps towards a novel vertebrate memory system that is not bound to the hippocampus. This has the potential to start a completely new avenue of research on the neural fundaments of memory.
If small brained, non-cortical pigeons would fulfil the same behavioural, electrophysiological, and imaging-based criteria that are typical for conscious experiences in humans and monkeys, we might be inclined to conceive consciousness as a mental state owned by many animals. This does not imply that pigeon consciousness “feels the same” as in humans. But it would mean that some conscious (for lack of a better word) self-experience of an organism, however simple, would be given to many, if not most animals. Such a data set would have an impact that goes much beyond cognitive neuroscience and should stir discussions from the humanities up to fields like animal ethics.
• What are the overall objectives?
In my project, I aim to implement three conceptual changes:
- To understand the neural basis of avian executive functions, we need to move from studying individual areas to cognition-specific pallial networks that together form the executive system. The activation of this system could then be accompanied by the deactivation of a bird-specific default mode network. To test these hypotheses, I will identify and analyze these network structures and their components using cognitive experiments, fMRI, electrophysiology and optogenetics.
- Key components of the avian hippocampus differ from those of mammals, so that avian memory may be less dependent on the hippocampus. I will use fMRI-based memory tasks and optogenetics to test whether birds have indeed evolved a completely different memory system.
- Consciousness, albeit in simpler forms, could be a mental state that many animals, including birds, possess. Here I will test whether neural correlates of consciousness, as shown in humans and monkeys, can also be demonstrated in pigeons using behavioral studies, fMRI and electrophysiology.
One important aspect of my ERC-project is the analysis of the avian hippocampus and its putative memory function. To this end, we conducted a detailed anatomical analysis of the intrahippocampal connectivity patterns in pigeons (Rook et al., 2023). Our techniques revealed a complex connectivity pattern along the hippocampal transverse axis that started in the dorsolateral hippocampus and continued to the dorsomedial subdivision, from where information was relayed to the triangular region either directly or indirectly via the V-shaped layers. The often-reciprocal connectivity along these subdivisions displayed an intriguing topographical arrangement such that two parallel pathways could be discerned along the ventrolateral (deep) and dorsomedial (superficial) aspects of the avian hippocampus. Overall, our findings provide an unprecedented, detailed description of avian intrahippocampal pathway and provide further support for the hypothesized homology of the lateral V-shape layer and the dorsomedial hippocampus with the dentate gyrus and Ammon’s horn of mammals, respectively.
A core technique of the project is fMRI in awake birds under ultrahigh magnetic fieldconditions. However, this is challenging due to strong local magnetic field inhomogeneities caused by air cavities in the avian skull. Therefore, we developed a two-segmented spin-echo echo-planar imaging (SE-EPI) sequence that covers the whole brain of awake pigeons. This sequence was applied to investigate sensory networks in awake pigeons and assessed the relative merits of this method in comparison with the classic single-shot RARE sequence. At the same imaging resolution but with a volume acquisition of 3 s versus 4 s for RARE, the two-segmented SE-EPI provided twice the strength of BOLD activity compared with the single-shot RARE sequence, while the image signal-to-noise ratio (SNR) and in particular the temporal SNR were very similar for the two sequences. In addition, the activation patterns in two-segmented SE-EPI data were more symmetric and larger than single-shot RARE results. Two-segmented SE-EPI thus represents a valid alternative to the RARE sequence in avian fMRI research since it yields more than twice the BOLD sensitivity per unit of time with much less energy deposition and better temporal resolution, particularly for event-related experiments. This will greatly benefit the project (Khodadadi et al., 2023).
One of our hypotheses was that the constitution of memory engrams go along with an activation of the avian hippocampus (or other structures). To this end, we studied filial imprinting in chicks by developing a ground-breaking non-invasive functional MRI technique for awake, newly hatched chicks that record whole-brain BOLD signal changes throughout imprinting experiments. Our findings identified potential long-term storage areas of imprinting memories across a neural network that included the hippocampal formation, providing strong support for its memory function in birds (Behroozi et al., in revision).
A key aim of this ERC was to identify cognitive networks. As a first approach, we studied dream patterns in pigeons within an ultrahigh magnetic force fMRI system. We show that REM sleep, a paradoxical state with wake-like brain activity, during which we experience our most vivid dreams, is accompanied in birds with the activation of networks involved in processing visual information, including optic flow during flight (Ungurean et al., 2023).
One of the aims of my ERC is to show consciousness in avian species that is not regarded as “feathered apes” like corvids. Many scientists ascribe the ability to recognize oneself in the mirror as a correlate of consciousness. Up to now, this ability was only shown in a few corvid species. We now used a completely novel approach by using roosters (which do not pass the usual mark-and-mirror test) under ecologically appropriate conditions: Roosters warn conspecifics when seeing an aerial predator, but not when alone. Exploiting this natural behavior, we tested individual roosters alone, with another male, or with a mirror while a hawk’s silhouette flew above them. Roosters mainly emitted alarm calls in the presence of another individual but not when alone or seeing themselves in the mirror. Thus, chickens possibly recognize their reflection as their own but only when tested under a condition in which they differentiate between alone or being accompanied by other chicken. This study shows how strikingly strong cognition is ecologically embedded (Hillemacher et al., 2023).
Using a similar ecological approach, we analyzed in greatest detail pigeons’ decision processes when being confronted with various options during a semi-natural foraging task. Our results indicate that hungry pigeons preferred to peck for delay reduction but did not work more for that option than for probability increase, which was the most profitable alternative and did not induce more pecking effort. These results cast a new light on current foraging theories and make it likely that avian decisions integrate many economic variables (Wittek et al., 2024).
In a major review (Güntürkün et al., 2024), we then outlined first pillars of a new theory why birds are smart. We propose four features that may be required for complex cognition: first, many associative pallial neurons, second, a prefrontal cortex (PFC)-like area, third, a dense dopaminergic innervation of association areas that feedback the outcome of own decisions, and fourth, dynamic neurophysiological fundaments withing the beta- and gamma-range for working memory. These four neural features have convergently evolved and may therefore represent ‘hard to replace’ mechanisms enabling complex cognition. Importantly, we could show that seemingly simple computational models can show that some of these properties, like large numbers of associative pallial neurons, rest on the power of associative learning—a process that is often underestimated in cognitive science (Wasserman et al., 2024).
A core technique of the project is fMRI in awake birds under ultrahigh magnetic fieldconditions. However, this is challenging due to strong local magnetic field inhomogeneities caused by air cavities in the avian skull. Therefore, we developed a two-segmented spin-echo echo-planar imaging (SE-EPI) sequence that covers the whole brain of awake pigeons. This sequence was applied to investigate sensory networks in awake pigeons and assessed the relative merits of this method in comparison with the classic single-shot RARE sequence. At the same imaging resolution but with a volume acquisition of 3 s versus 4 s for RARE, the two-segmented SE-EPI provided twice the strength of BOLD activity compared with the single-shot RARE sequence, while the image signal-to-noise ratio (SNR) and in particular the temporal SNR were very similar for the two sequences. In addition, the activation patterns in two-segmented SE-EPI data were more symmetric and larger than single-shot RARE results. Two-segmented SE-EPI thus represents a valid alternative to the RARE sequence in avian fMRI research since it yields more than twice the BOLD sensitivity per unit of time with much less energy deposition and better temporal resolution, particularly for event-related experiments. This will greatly benefit the project (Khodadadi et al., 2023).
One of our hypotheses was that the constitution of memory engrams go along with an activation of the avian hippocampus (or other structures). To this end, we studied filial imprinting in chicks by developing a ground-breaking non-invasive functional MRI technique for awake, newly hatched chicks that record whole-brain BOLD signal changes throughout imprinting experiments. Our findings identified potential long-term storage areas of imprinting memories across a neural network that included the hippocampal formation, providing strong support for its memory function in birds (Behroozi et al., in revision).
A key aim of this ERC was to identify cognitive networks. As a first approach, we studied dream patterns in pigeons within an ultrahigh magnetic force fMRI system. We show that REM sleep, a paradoxical state with wake-like brain activity, during which we experience our most vivid dreams, is accompanied in birds with the activation of networks involved in processing visual information, including optic flow during flight (Ungurean et al., 2023).
One of the aims of my ERC is to show consciousness in avian species that is not regarded as “feathered apes” like corvids. Many scientists ascribe the ability to recognize oneself in the mirror as a correlate of consciousness. Up to now, this ability was only shown in a few corvid species. We now used a completely novel approach by using roosters (which do not pass the usual mark-and-mirror test) under ecologically appropriate conditions: Roosters warn conspecifics when seeing an aerial predator, but not when alone. Exploiting this natural behavior, we tested individual roosters alone, with another male, or with a mirror while a hawk’s silhouette flew above them. Roosters mainly emitted alarm calls in the presence of another individual but not when alone or seeing themselves in the mirror. Thus, chickens possibly recognize their reflection as their own but only when tested under a condition in which they differentiate between alone or being accompanied by other chicken. This study shows how strikingly strong cognition is ecologically embedded (Hillemacher et al., 2023).
Using a similar ecological approach, we analyzed in greatest detail pigeons’ decision processes when being confronted with various options during a semi-natural foraging task. Our results indicate that hungry pigeons preferred to peck for delay reduction but did not work more for that option than for probability increase, which was the most profitable alternative and did not induce more pecking effort. These results cast a new light on current foraging theories and make it likely that avian decisions integrate many economic variables (Wittek et al., 2024).
In a major review (Güntürkün et al., 2024), we then outlined first pillars of a new theory why birds are smart. We propose four features that may be required for complex cognition: first, many associative pallial neurons, second, a prefrontal cortex (PFC)-like area, third, a dense dopaminergic innervation of association areas that feedback the outcome of own decisions, and fourth, dynamic neurophysiological fundaments withing the beta- and gamma-range for working memory. These four neural features have convergently evolved and may therefore represent ‘hard to replace’ mechanisms enabling complex cognition. Importantly, we could show that seemingly simple computational models can show that some of these properties, like large numbers of associative pallial neurons, rest on the power of associative learning—a process that is often underestimated in cognitive science (Wasserman et al., 2024).
Up to now, all components of the project progress as planned. As a result, I plan to insert one addition that resulted as a spin-off from the Ungurean et al. (2023) paper. I plan to “read a pigeon’s mind” by running an fMRI-study in which I confront the pigeons with sequences of images within the scanner. While seeing an object, we can identify the voxel pattern of brain activities by using multi voxel pattern analysis (MVPA). Thus, for each object, we then have a corresponding pattern. Once we have that, we can identify the MVPA-pattern for the object that the animal anticipates (thinks about) to occur within the next seconds on the screen. Once we have that, we could go on planned many further studies on the AVIAN MIND.