Purchase this article with an account.
Moshe Bar, Elissa Aminoff; Contextual processing of visual objects in the brain. Journal of Vision 2002;2(7):410. doi: 10.1167/2.7.410.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Objects that share the same context tend to appear together in our environment. Is this frequent co-occurrence manifested in the organization of their representation in the brain? Given how little is known about the neuronal analysis of context, the first step in addressing this question is to define the cortical network involved. To elicit maximal contextual activation, we first identified for each of the 40 contexts in the experiment a “key object” that is most commonly associated with that specific context (e.g., a flower for a garden, a hardhat for a construction site). In addition, we identified another set of objects that are not associated with any specific context (e.g., a rope). Average reaction time for recognizing the objects in both sets was equal. We then used fMRI to scan six subjects while they were engaged in recognizing those individual objects (3T magnet; TR=2sec; 33 slices, 3mm each). The cortical activation elicited by highly contextual objects (“key objects”) was compared with the cortical activation elicited by “non-contextual” objects. The resulting differential activation concentrated in the temporal cortex, primarily in the collateral sulcus and extended into the parahippocampal gyrus. This focus, which was exceptional in its extent, significance, and consistency across subjects, is suggested to play a central role in the representation and processing of context. What is the specific function associated with this focused activity? One possible explanation is that it reflects the co-activation of multiple objects that share the same context, triggered by the recognition of a “key object.” Such contextual representations may include information about typical members of each context as well as the typical spatial relations between those objects (schema or “context frames”). Alternatively, this activation may be a manifestation of semantic and abstract information associated with the specific contexts. We will contrast these accounts.
This PDF is available to Subscribers Only