Portable, non-destructive colorimetry and visible reflectance spectroscopy paired with machine learning can classify experimentally heat-treated silcrete from three South African sources

Abstract

The objective of this study was to determine if visible reflectance spectroscopy and quantitative colorimetry represent viable approaches to classifying the heat treatment state of silcrete. Silcrete is a soil duricrust that has been used as toolstone since at least the Middle Stone Age. The ancient practice of heat treating silcrete prior to knapping is of considerable interest to paleolithic archaeologists because of its implications for early modern human complex cognition generally and the ability to manipulate the material properties of stone specifically. Here, we demonstrate that our quantitative, non-invasive, and portable approach to measuring color, used in conjunction with k-Nearest Neighbors “lazy” machine learning, is a highly promising method for heat treatment detection. Traditional, expert human analyst approaches typically rely upon subjective assessments of color and luster and comparison to experimental reference collections. This strongly visual method can prove quite accurate, but difficult to reproduce between different analysts. In this work, we measured percent reflectance for the visible spectrum (1018 variables) and standardized color values (CIEL*a*b*) in unheated and experimentally heat-treated silcrete specimens from three sources in South Africa. k-NN classification proved highly effective with both the spectroscopy and colorimetry data sets. An important innovation was using the heat treatment state predicted by the k-NN model for the majority of replicate observations of a single specimen to predict the heat treatment state for the specimen overall. When this majority voting approach was applied to the 746 individual observations in this study, associated with 94 discrete silcrete flakes, both spectroscopy and colorimetry k-NN models yielded 0% test set misclassification rates at the specimen level.

An integrated approach to reconstruct the role of the heat treatment within the reduction sequence of chert artefacts: The case of the early Neolithic site of Lugo di Grezzana (north‐eastern Italy)

“…This study proposes a methodological advancement for the investigation of intentional thermal treatment of archaeological chert artefacts. The use of a multi‐analytical approach proved to be particularly valuable for confirming what was recently stressed by other authors (Murray et al, 2022). Both density measurements and FT‐IR reflectance analyses allow to test any kind of archaeological sample avoiding any selection due to size and morphology.…”

Three South African silcrete sources can be identified regardless of heat treatment using solution ICP-MS and LA-ICP-MS

“…The heat treatment process used for this research is described in detail by Murray et al (2020Murray et al ( , 2022 and Oestmo (2017), so only summary information is provided here. Silcrete nodules from each source were cut to create paired 7 x 7 x 15 cm blocks to get one heated and one unheated block from the same piece of rock.…”

“…Between samples, we cleaned the vials by milling ashed quartz (i.e., ne sand with all organics removed) for 2 minutes and wiping down the vial using distilled water and ethanol. Prior to destruction, each ake was scanned with a Variable Inc. Spectro 1 Pro hand-held colorimeter to record quantitative color measurements (Murray et al, 2022). Further, we used a silicone peeling compound to permanently record the surface roughness of these akes (Murray et al, 2020).…”