Travel Routes and Spatial Abilities in Wild Chacma Baboons (Papio ursinus)

Abstract

The primary objective of this research was to give insight into the spatial cognitive abilities of chacma baboons (Papio ursinus) and to address the question whether chacma baboons internally represent spatial information of large-scale space in the form of a so-called topological map or a Euclidean map. Navigating the environment using a topological map envisions that animals acquire, remember and integrate a set of interconnected pathways or route segments that are linked by frequently used landmarks or nodes, at which animals make travel decisions. When animals navigate using a Euclidean map, animals encode information in the form of true angles and distances in order to compute novel routes or shortcuts to reach out of view goals. Although findings of repeatedly used travel routes are generally considered evidence that animals possess topological-based spatial awareness, it is not necessarily evidence that they navigate (solely) using a topological map or lack complete Euclidean spatial representation. Therefore, three predictions from the hypothesised use of a topological map and Euclidean map were tested to distinguish between them. It was investigated whether there was a difference in travel linearity between the core area and the periphery of the home range, whether travel goals were approached from all directions or from one (or a few) distinct directions using the same approach routes and lastly, whether there was a difference between the initial leaving direction from a travel goal and the general direction towards the next goal. Data were collected during a 19-month period (04/2007-11/2008) at Lajuma research centre in the Soutpansberg (Limpopo Province, South Africa). A group of baboons were followed from their morning sleeping site to their evening sleeping site for 234 days, during which location records, behavioural data and important resource data were recorded. A statistical procedure termed the change-point test (CPT) was employed to identify locations at which baboons started orienting towards a goal and baboons showed goal-directed travel towards identified travel goals. Subsequently, hotspot analysis was employed to delineate clusters of such change-points, termed ‘decision hotspots’. Decision hotspots coincided with highly valuable resources, towards which baboons showed significantly faster travel. It thus seemed that they ‘knew’ when they were nearing their goals and adapted their speed accordingly. Decision hotspots were also located at navigational landmarks that delineated a network of repeatedly used travel routes characteristic of a topological map. Therewith, this method reveals an important utility to the study of decision-making by allowing a range of sites to be selected for detailed observations, which were previously limited to sleeping sites or ‘stop’ sites, which would be impossible if the decision hotspots had not been previously identified. Furthermore, baboons travelled as efficiently in the periphery as in the core area of their home range, which was suggested to be more consistent with Euclidean spatial awareness. However, comparatively low travel linearity throughout the home range revealed it is more likely that the baboons accumulated a similar knowledge of the periphery as of the core area, which allowed them to navigate with a similar efficiently through both areas. The mountainous terrain at the study site provided ample prominent landmarks to aid the baboons in navigation and allowed baboons to initiate navigation to a travel goal with the same direction as when they reached that goal. Baboons did not approach travel goals from all directions, but instead they approached their goals from the same direction(s). In conclusion, the findings of this research are more consistent with the use of a topological spatial representation of large scale space, where landmarks aid baboons to navigate efficiently through large scale space. A review of the literature shows that until date, evidence for the existence of Euclidean spatial representation in both animals and humans is extremely limited and often unconvincing. It is likely that a high level of experimental control is necessary to unambiguously demonstrate the existence of Euclidean spatial awareness in the future.