Philipp Haueis | What the brain does by itself A patchwork approach to endogenous brain activity in resting state functional connectivity research

One way in which philosophy of neuroscience can contribute to ongoing research

is to articulate new frameworks that assess the status of central concepts and which open up

novel experimental possibilities. In this talk I adopt philosophical work on the patchwork

structure of concepts (Wilson 2006) and causal specificity (Woodward 2010) to assess the

concept of “functional connectivity” in resting state research. This concept refers to coordinated

endogenous brain activity that occurs in the absence of a task, but the functional role of this

activity remains debated. The direct role view posits that this activity is operative in information

processing mechanisms because it has specific effects on capacities such as motor control (Fox

et al. 2007) or mind-wandering (Smallwood et al. 2016). In contrast, the enabling condition

view holds that endogenous activity maintains the structural integrity of brain systems, which

is a necessary prerequisite for information processing to occur (Larsen-Prior et al. 2009, Raichle

2015). In this talk, I argue that there is no need to choose between the direct role and enabling

condition view. Instead, we should adopt a patchwork approach to study endogenous activity

with different functional roles. The patchwork approach holds that “functional connectivity” in

resting state studies can refer to different kinds of endogenous brain activity, depending on the

technique and analysis we use in an experiment (e.g., fMRI, EEG, electrodes in animal studies,

or static vs. dynamic connectivity analyses). The patchwork approach sorts these activities with

regard to their causal specificity in cognition and behavior. Under some conditions, resting state

studies measure endogenous activity with cognitive roles, such as information integration via

the posterior cingulate cortex during mind-wandering (Smallwood et al. 2016). In other

conditions, they may track activity that enables cognition, such as energetic resources

(Vaishnavi et al. 2010) or fluctuations of the brain’s global waste disposal system (Kiviniemi

et al. 2016). Yet in other conditions, they can track activity with intermediate roles such as

arousal (Chang et al. 2016), which modulates the performance of multiple cognitive tasks at

once. The patchwork approach helps us to keep these operative, modulatory or enabling roles

in view, whereas the direct role or enabling condition views alone do not. The patchwork

approach therefore helps to formulate new experiments that investigate the relation between

information processing and maintenance functions of endogenous activity, or dissect

frequency-specific contributions to the signals measured during the resting state.