Cognitive performance and specific aspects of language processing are associated with Oldowan-like chert flaking and retouch
A vast amount of literature suggests a co-evolutionary relationship between Palaeolithic stone toolmaking, and cognition and specifically language. However, empirical data remain limited to indirect findings of neurophysiological studies. Furthermore, most Oldowan studies have used chert and have not investigated retouch, even though quartz and lava were predominant raw materials during periods of chert unavailability, and even though chert was disproportionately more frequently used for
... compared to other raw materials during periods of chert availability, at least in the Olduvai Gorge. The study recruited 13 young adults with no prior experience in knapping. Subjects were taught by an experienced knapper to produce quartz choppers and chert sidescrapers in either a verbal or gestural condition. Two raters rated on a 5-point scale the subjects' performances on specific steps of the two stone toolmaking tasks. In a post-experimental interview, subjects stated which aspects of the tasks they preferred or disfavored. Subjects also performed on a neuropsychological battery encompassing visuospatial, executive functioning, and linguistic tasks. Given the small sample size, the results should be regarded as exploratory and preliminary. Our results are further limited to the early acquisition phase and may not reflect processes in modern experienced knappers. Descriptive data suggested better performance across all stone toolmaking variables in the verbal compared to gestural condition, but only flake quality on the sidescraper task was significantly different between groups. Analyses of the stone toolmaking variables suggested subjects perceived quartz and chert flaking very differently. Correlational and other analyses suggested that quartz chopper manufacture was not associated with cognitive performance. Conversely, chert flaking and retouch were strongly associated with visuospatial working memory, showing that subjects with a higher memory span produced better chert flakes and retouch. Retouch only was moderately associated with executive functioning measures, showing subjects who made fewer errors on the tasks were better on retouch. Specific aspects of chert flaking were also associated with verbal fluency performance, showing, among others, moderate and strong positive associations with the productivity and rate of production of syntactically transitive verbs on action fluency. Evolutionary implications can be drawn from our research only if we controversially assume similar results would have been obtained had we tested early hominins and not modern humans. Following this axiom, our results suggest that Oldowan hominins relied on modern-like visuospatial working memory during chert flaking and retouch, and, to a lesser degree, modern-like executive functioning during chert retouch. This is contrary to previous Oldowan studies suggesting no involvement of executive functioning during Oldowan-like flaking. Results from the linguistic tasks controversially suggest that some of the prerequisites for aspects of action language and syntactic transitivity (verb-object phrases) in modern humans were to some degree present in Oldowan hominin populations. Because Olduvai Gorge hominins readily incorporated chert for stone toolmaking in periods of chert availability, our results suggest that these cognitive capacities were phylogenetically not related to chert knapping. Finally, we propose that the quality of performance on Oldowan flaking and retouch may not reflect the full level of cognitive capacities of Oldowan populations. We provide the first direct evidence for an association between Palaeolithic stone toolmaking and cognitive performance in modern humans, while previous studies have inferred cognitive processes from neuroimaging data. We also provide the first direct evidence for an association between Palaeolithic stone toolmaking, and action language and simple syntactic transitivity in modern humans.