No individual differences in figure recognition and visual attention patterns during the observation of Palaeolithic art

15th ESHE Meeting - European Society for the study of Human Evolution

Paris. France
Conference. 2025

Art visual processing is a complex cognitive function that involves perception, attention and decision-making. In particular, Palaeolithic art represents a structured form of graphic communication, intended to be perceived and interpreted by members of the original human group, with the potential for a lasting impact beyond the moment of creation (Hodgson, 2003). The first stage of this visual system is characterized by exploration through rapid eye movements, known as saccades, which are produced between small pauses called fixations. These eye movements can be recorded by eye tracking technology and provide insights into attentional patterns and information processing (Carrasco, 2011). Moreover, according to active inference models of cognition, eye movements play a key role in minimizing prediction errors generated by the brain through comparisons between perceived input and prior experiences (Friston, 2009). From this perspective, vision operates as a predictive system that manages both external salience factors and internally generated information.

In this study, we employ eye-tracking to investigate visual perception during the observation of Palaeolithic rock art, focusing on the behavioural and biological mechanisms involved in figure recognition and attentional allocation. Participants were instructed to freely observe a series of photographs of zoomorphic figures from several Cantabrian seaboard caves for a duration of 10 seconds while their eye movements were recorded using a screen-based eye tracker. After finished the observation, participants were asked if they recognised the animals depicted. The areas of greatest visual saliency and therefore likely to attract attention were calculated by graph-based visual saliency (GBVS) computation algorithm. Then, different areas of interest (AOIs) were defined based on these most computed salient locations (saliency AOI), other elements in the cave such as fissures, calcite formations and other changes in the morphology of the support (cave AOI), as well as the depiction. Furthermore, some AOIs were defined based on the main anatomical regions of the figures, namely head, anterior, body and posterior. We computed the dwell time (in milliseconds) relative to the size of each AOI (in pixels) as a measure of fixation density over an area (Silva-Gago et al., 2022). This variable (relative dwell time, DT_REL) standardizes the amount of attention directed to each AOI regardless of their size. Our results suggest that attention was predominantly directed to the depicted figures, in particular to their heads, rather than to other elements of the visual field, regardless of whether or not the participant recognised the depicted animal. Even when the animal was not identified, gaze and attention consistently focused on the most informative areas of the panel, indicating that the painting guides visual exploration. Additionally, no significant differences were observed based on participants’ gender or prior archaeological and domain-specific knowledge, at least with respect to the variables and depictions employed in this study. To conclude, we propose eye tracking as a tool to study the perception of rock art from a cognitive point of view, revealing that the paintings guide the observer and create common patterns of observation which facilitate the identification of the animal depicted (Meyering et al., 2020).

Eye-tracking. Perception. Rock art. Vision. Cognitive archaeology.