Periodic Reporting for period 1 - ViSyRelPer (Perceptual foundations of relational thinking)
Reporting period: 2023-05-01 to 2025-04-30
The ability to represent relations, the properties holding between things, is regarded as a distinctive feature that can account for the complexity and uniqueness of the human mind. Yet, we do not know how relational representations come about in the human mind/brain. Are relations represented discretely and explicitly, separately from the objects participating in them? Or are they represented as basic properties of the objects in the scope of the relation?
Relational representations have primarily been studied in the context of higher-level cognitive functions such as language and reasoning. Only recently, scholars have begun to investigate their perceptual underpinnings, yielding a new research field, which exploits the constructs and methods of vision science to investigate whether and how we see (i.e. encode) relations holding between objects, just like we see features such as colour, motion, shape and so on. This initial work on relation perception has shown that the early stages of the representation of relations between objects in visual scenes are encoded in visual areas. Visual areas would host specialized structures for representing objects as well as their relations, such as support (A is on B), containment (A is in B), and other relational aspect of the world such as causality (A hits B that starts moving) and social interaction (A does something to/with B).
The working hypothesis of ViSyRelPer is that the visual system computes high-level relations, categorizing configurations of multiple objects into distinct relational types, independently from the objects that constitute the relation. On this hypothesis, in vision, depictions of a "tiger in a bath" and "a flower in a vase" will be recognized as instances of the same relation “in” (containment), despite their perceptual dissimilarities. In addressing this hypothesis, ViSyRelPer asks how far vision can go in representing the visual world, particularly whether it can afford high-level representations of relations that generalize across different situations that do not share the same visual featues. The overarching goal of ViSyRelPer is to study the role of visual perception in the representation of relations. This research will contribute to determine the computational goals of human vision, and to develop a model that accounts for how relational contents come about in the human mind/brain.
Relational representations have primarily been studied in the context of higher-level cognitive functions such as language and reasoning. Only recently, scholars have begun to investigate their perceptual underpinnings, yielding a new research field, which exploits the constructs and methods of vision science to investigate whether and how we see (i.e. encode) relations holding between objects, just like we see features such as colour, motion, shape and so on. This initial work on relation perception has shown that the early stages of the representation of relations between objects in visual scenes are encoded in visual areas. Visual areas would host specialized structures for representing objects as well as their relations, such as support (A is on B), containment (A is in B), and other relational aspect of the world such as causality (A hits B that starts moving) and social interaction (A does something to/with B).
The working hypothesis of ViSyRelPer is that the visual system computes high-level relations, categorizing configurations of multiple objects into distinct relational types, independently from the objects that constitute the relation. On this hypothesis, in vision, depictions of a "tiger in a bath" and "a flower in a vase" will be recognized as instances of the same relation “in” (containment), despite their perceptual dissimilarities. In addressing this hypothesis, ViSyRelPer asks how far vision can go in representing the visual world, particularly whether it can afford high-level representations of relations that generalize across different situations that do not share the same visual featues. The overarching goal of ViSyRelPer is to study the role of visual perception in the representation of relations. This research will contribute to determine the computational goals of human vision, and to develop a model that accounts for how relational contents come about in the human mind/brain.
In the ViSyRelPer project, substantial progress has been made toward understanding how the human visual system encodes abstract relations between objects in a scene — a foundational cognitive ability that underpins reasoning, communication, and social understanding.
The project began by developing a novel set of visual stimuli that depict relations such as containment, support, and interaction between objects or agents. Unlike earlier studies using language or static images, ViSyRelPer employs short, movie-like sequences to explore whether the brain can automatically and rapidly detect relational patterns, independently of the specific objects involved.
To test this, I designed a series of EEG experiments using the Fast Periodic Visual Stimulation (FPVS) technique — an innovative method that tracks the brain’s automatic response to visual categories. Stimuli were presented in rapid sequences where object relations were either held constant or systematically varied. This allowed us to measure how the brain distinguishes abstract relations based solely on visual input, and how quickly these distinctions emerge.
EEG data collection for these experiments is complete. Results suggest that the brain shows selective and robust responses to visual relations, with distinct neural signatures emerging in early visual and parietal regions — supporting the idea that relational encoding occurs early in the perceptual process, not only at later stages involving reasoning or language.
A second set of experiments investigated the generalization of relational perception: for example, whether the brain treats “a cat under a table” and “a book under a lamp” as instances of the same relation. Data collection is finalized, and the results indicate that, to a certain extent, the brain can generalize across visually dissimilar configurations — suggesting an object-independent coding of relations.
In a follow-up EEG study, we tested whether the response to relations like “containment” and “support” is fully abstract. Results showed that the brain’s response is not entirely independent of object type; for example, the nature of the container or support object influences the response, pointing to limits in abstraction.
To extend these findings, I developed a new set of abstract stimuli that rely on non-meaningful shapes, allowing us to test relational perception beyond familiar objects.
In parallel, the project has laid the foundation for a future work package exploring whether relational perception differs in individuals with autism spectrum conditions, with preparatory work underway.
Overall, ViSyRelPer has delivered key scientific outputs: a novel experimental paradigm, new insights into how and where the brain encodes relational information, and the groundwork for translational applications in neurodevelopmental research. These findings contribute to theoretical models of visual cognition and may ultimately inform clinical approaches to diagnosing or supporting individuals with atypical perceptual or cognitive profiles.
The project began by developing a novel set of visual stimuli that depict relations such as containment, support, and interaction between objects or agents. Unlike earlier studies using language or static images, ViSyRelPer employs short, movie-like sequences to explore whether the brain can automatically and rapidly detect relational patterns, independently of the specific objects involved.
To test this, I designed a series of EEG experiments using the Fast Periodic Visual Stimulation (FPVS) technique — an innovative method that tracks the brain’s automatic response to visual categories. Stimuli were presented in rapid sequences where object relations were either held constant or systematically varied. This allowed us to measure how the brain distinguishes abstract relations based solely on visual input, and how quickly these distinctions emerge.
EEG data collection for these experiments is complete. Results suggest that the brain shows selective and robust responses to visual relations, with distinct neural signatures emerging in early visual and parietal regions — supporting the idea that relational encoding occurs early in the perceptual process, not only at later stages involving reasoning or language.
A second set of experiments investigated the generalization of relational perception: for example, whether the brain treats “a cat under a table” and “a book under a lamp” as instances of the same relation. Data collection is finalized, and the results indicate that, to a certain extent, the brain can generalize across visually dissimilar configurations — suggesting an object-independent coding of relations.
In a follow-up EEG study, we tested whether the response to relations like “containment” and “support” is fully abstract. Results showed that the brain’s response is not entirely independent of object type; for example, the nature of the container or support object influences the response, pointing to limits in abstraction.
To extend these findings, I developed a new set of abstract stimuli that rely on non-meaningful shapes, allowing us to test relational perception beyond familiar objects.
In parallel, the project has laid the foundation for a future work package exploring whether relational perception differs in individuals with autism spectrum conditions, with preparatory work underway.
Overall, ViSyRelPer has delivered key scientific outputs: a novel experimental paradigm, new insights into how and where the brain encodes relational information, and the groundwork for translational applications in neurodevelopmental research. These findings contribute to theoretical models of visual cognition and may ultimately inform clinical approaches to diagnosing or supporting individuals with atypical perceptual or cognitive profiles.
Results Beyond the State of the Art
The ViSyRelPer project has generated new evidence that the human visual system can encode abstract relations between objects — such as containment, support, and interaction — automatically and independently of the specific objects involved. This challenges traditional models that attribute relational reasoning primarily to higher-level, language-based cognitive processes, and positions visual perception itself as a key player in abstract cognition.
A key scientific advance lies in demonstrating that the brain distinguishes relational structures rapidly and automatically through specific neural signatures measurable via EEG, using the Fast Periodic Visual Stimulation (FPVS) technique. Prior to this project, relational processing was mostly studied in linguistic or symbolic domains. ViSyRelPer extends this understanding by providing the first systematic evidence that relational encoding begins early in the visual system, and that these relations can be generalized across different perceptual inputs.
Furthermore, the development of new dynamic visual stimuli — both with meaningful objects and abstract shapes — sets a methodological benchmark for future studies in cognitive neuroscience and developmental psychology. These tools will be made openly available, promoting further replication and innovation in the field.
Another important result is the identification of limits in relational abstraction: although some generalization occurs, the encoding of a relation is still partly influenced by the type of object involved (e.g. container vs. supported item). This nuance is critical for refining theoretical models of perception and cognition.
To ensure further uptake and success, several avenues should be pursued:
• Further research: Cross-cultural studies, developmental trajectories (e.g. in children), and clinical populations (e.g. ASD) should be investigated
• Data and tool sharing: The open distribution of stimuli and code will foster broader scientific collaboration and reproducibility
• Communication and public engagement
The ViSyRelPer project has generated new evidence that the human visual system can encode abstract relations between objects — such as containment, support, and interaction — automatically and independently of the specific objects involved. This challenges traditional models that attribute relational reasoning primarily to higher-level, language-based cognitive processes, and positions visual perception itself as a key player in abstract cognition.
A key scientific advance lies in demonstrating that the brain distinguishes relational structures rapidly and automatically through specific neural signatures measurable via EEG, using the Fast Periodic Visual Stimulation (FPVS) technique. Prior to this project, relational processing was mostly studied in linguistic or symbolic domains. ViSyRelPer extends this understanding by providing the first systematic evidence that relational encoding begins early in the visual system, and that these relations can be generalized across different perceptual inputs.
Furthermore, the development of new dynamic visual stimuli — both with meaningful objects and abstract shapes — sets a methodological benchmark for future studies in cognitive neuroscience and developmental psychology. These tools will be made openly available, promoting further replication and innovation in the field.
Another important result is the identification of limits in relational abstraction: although some generalization occurs, the encoding of a relation is still partly influenced by the type of object involved (e.g. container vs. supported item). This nuance is critical for refining theoretical models of perception and cognition.
To ensure further uptake and success, several avenues should be pursued:
• Further research: Cross-cultural studies, developmental trajectories (e.g. in children), and clinical populations (e.g. ASD) should be investigated
• Data and tool sharing: The open distribution of stimuli and code will foster broader scientific collaboration and reproducibility
• Communication and public engagement