Testing and training geometrical knowledge in baboons through a conceptual categorization task
Sensitivity to geometric regularity has been proposed as a signature of human cognitive singularity. Baboons can detect an intruder shape among quadrilaterals but, unlike humans, show no advantage for regular shapes such as squares and rectangles, suggesting they are not sensitive to geometric regularity. Here we asked whether, with symbolic training, baboons may show an enhanced attention to…
Recent research suggests that baboons possess a sensitivity to geometric regularity, a cognitive ability that sets humans apart. However, the results of a new study indicate that, despite being able to detect irregular shapes among quadrilaterals, baboons do not exhibit an enhanced attention to geometric features through symbolic training.
The study aimed to determine if baboons could learn to categorize stimuli based on geometric properties, such as perpendicularity, parallelism, and length equality, when exposed to new generalization stimuli.
Symbolic match-to-sample tasks were employed to train both baboons and human controls to categorize stimuli as belonging to a given concept (A) or not (not-A) across three geometric domains and two non-geometric domains. The baboons demonstrated the ability to learn and generalize across all categories, except for the length equality domain. This finding indicates that baboons can extract geometric properties and link them to arbitrary symbols.
Interestingly, the explicit geometry training did not lead to the emergence of human-like performance in the quadrilateral intruder task. While the baboons' overall performance improved uniformly across all shapes, the geometric regularity effect, a key characteristic of human cognition, failed to materialize. This result suggests that, in baboons, recognizing individual geometric properties alone is not sufficient to trigger the compositional perception of shapes, a defining feature of human cognitive abilities.
Written by urgent.news from bioRxiv's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.