Over the past few decades, research has identified two key frames of reference: the egocentric frame, which encodes object locations relative to the viewer, and the allocentric frame, which encodes the intrinsic coordinates of objects. The egocentric frame of reference has been found to emerge from a bilateral frontoparietal network, while the allocentric frame of reference is thought to recruit areas within the ventral occipitotemporal cortex. However, our understanding of the specific contributions of individual ventral regions to allocentric processing remains limited. Recent studies have identified a novel attentional area in the ventral stream - the posterior inferotemporal (PIT+) area. This area has been shown to have greater functional MRI (fMRI) activation during the performance of some allocentric tasks. Yet, whether and how the role of PIT+ may extend to novel allocentric tasks, underpin allocentric neglect, a spatial attention deficit following brain lesions, and may be causally involved in allocentric processing remain unknown. Crucially, due to a lack of reliable tests, allocentric neglect is rarely assessed in clinical settings, highlighting the need for new sensitive tools to be developed.
To this end, we conducted three experiments using a novel battery of tasks to investigate allocentric processing in healthy subjects, detect allocentric deficits in stroke patients, and explore causal brain-behavior relationships. In the first experiment, we used fMRI with healthy participants to measure PIT+ activity during a reference frame discrimination task comparing egocentric, allocentric, and color control conditions. In the second experiment, we tested whether our task could sensitively detect allocentric neglect in stroke patients. In the third, we used Transcranial Magnetic Stimulation (TMS) to assess whether stimulating PIT+ would alter behavior during allocentric processing.
Our results confirm that PIT+ is significantly activated during both allocentric and egocentric processing. However, its activation was not stronger in the allocentric condition of the new task we developed, mostly because it required alternative strategies to solve the task. We also found that specific features of the new battery were sensitive enough to detect stroke patients with allocentric neglect. Finally, we succeeded in stimulating PIT+ for the first time and observed some preliminary effects. Yet, to determine whether PIT+ is causally linked to allocentric reasoning, a much larger sample size would be needed. In conclusion, these findings further support PIT+’s role in spatial representations and highlight how stimulus properties, task difficulty, and stimulus design can influence its engagement.
