Understanding both the communicative and cognitive roles of language, the degree to which language facilitates particular cognitive activities, and what features of what we consider to be “normal” human cognition are enabled or assisted by language are all necessary to comprehend why language arose in the hominid lineage.
Language learning and usage are thought to be feasible because humans possess specific cognitive abilities, such as systematicity and symbolic thinking, according to a prevalent view in cognitive science. An alternative perspective is that acquiring and using a natural language alters human intellect playing an important role in symbolic reasoning.
Geometric reasoning, which is said to draw on a universal and language-independent human capacity, is examined in Gary Lupian’s et al work as a powerful test for the involvement of language in reasoning. A study by Dehaene et al. (2006) demonstrating a strong correlation in performance on an odd-one-out geometric reasoning task between educated Americans and the Munduruk, an Amazonian indigenous people without formal education and who lack vocabulary for describing the geometric relations in question, provides important evidence in favor of the universality and language-independence of geometric reasoning.
The team tested for the causal involvement of language in geometric reasoning in six ways:
- if naming geometric relations improved reasoning performance in English-speaking adults: resulting in improving performance
- if interfering with language-impaired performance: resulting in impairing performance
- if naming the relations in English selectively predicts English-speaker performance: resulting in significant correlations between performance and nameability
- if naming predicted the detrimental impact of verbal interference: resulting in an increased dependence on nameability
- how the performance of congenitally deaf children residing in a Chinese special school for the deaf who were deprived of normal language input for most of their childhood compares with normal language input and Chinese adults: resulting in lower scores from the children with language impairment
- collecting data on the task from the Shawi, an indigenous group of horticulturalists traders from Northwestern Amazonia who speak a Kawapanan language
Our results replicate Dehaene et al’s finding of substantial correlations in performance even among these very disparate populations. However, these correlations appear to reflect shared visual processing mechanisms rather than shared geometric reasoning abilities.
The results provide an argument for a geometric reasoning facilitated by language, with the construction of a more categorical hypothesis space. The capacity to identify things and their relationships (e.g., square, parallel, right-angle) when faced with a variety of items offers an efficient technique to abstract away perceptual aspects that might otherwise predominate the categorization response. The ramifications of this work go beyond geometry, demonstrating how linguistic enculturation and active language usage are necessary for the development of cognitive processes that are commonly believed to be nonlinguistic.
Core knowledge or language augmented cognition? The case of geometric reasoning. Lupyan, G., Wendorf, A., Berscia, L. M., & Paul, J.
Published: 2018, Proceedings of the 12th International Conference on the Evolution of Language (EVOLANG XII) (pp. 252-254). Toruń, Poland: NCU Press.