It has long been recognized that a simple theoretical solution to this instability problem is to endow neurons with homeostatic plasticity mechanisms that keep neuronal firing HTS assay rates within a set point range (Miller and MacKay, 1994), but whether neuronal firing in the intact CNS is homeostatically regulated remains a critical and untested prediction of the neuronal homeostasis hypothesis. Here we used a monocular visual deprivation (MD) paradigm to ask whether neurons within primary visual cortex (V1) homeostatically regulate their firing rates back to a set point value during a prolonged sensory perturbation. Visual deprivation paradigms followed by ex vivo

measurements in V1 have identified several forms of Hebbian and homeostatic plasticity that are expressed in a layer- and cell-type-specific manner

and are activated with distinct temporal profiles (Kirkwood et al., 1996, Rittenhouse et al., 1999, Desai et al., 2002, Maffei et al., LY2835219 cost 2006, Maffei et al., 2010, Maffei and Turrigiano, 2008, Kaneko et al., 2008 and Lambo and Turrigiano, 2013). Because of this complexity, the net effect of visual deprivation on activity within V1 is difficult to predict based on ex vivo measurements alone. Attempts to measure activity homeostasis in the intact visual cortex have not so far been definitive; in vivo calcium or intrinsic signal imaging in anesthetized animals revealed that MD first reduced and then increased visual drive (Mrsic-Flogel et al., 2007 and Kaneko et al., 2008), but average visual drive was not well conserved during this process (Mrsic-Flogel et al., 2007). Visually driven activity in anesthetized animals may not be the best probe for firing rate homeostasis for a number of reasons; most critically, because

homeostatic plasticity operates over a timescale of many hours (Turrigiano, 2008), it presumably normalizes some metric of average activity that will include both visually driven and spontaneous (or internally driven) spikes. We therefore set out to chronically monitor firing in V1 of freely viewing almost and behaving rodents over many days so that we could sample all spikes regardless of origin and directly determine whether average V1 firing rates are restored to baseline during MD. We used a classic MD paradigm (lid suture) to perturb visual drive in juvenile rats during a developmental period (postnatal days 27–32 [P27–P32]), when this perturbation is known to induce both Hebbian and homeostatic forms of plasticity within V1 (Smith et al., 2009, Turrigiano, 2011 and Levelt and Hübener, 2012). We obtained chronic multielectrode recordings as described (Jones et al., 2007, Sadacca et al., 2012 and Piette et al., 2012) from both hemispheres of monocular V1 in freely behaving animals, recorded several hours of activity during the same circadian period each day for 9 days, and separated units into putative PV+ fast-spiking basket cells (pFS) or regular-spiking units (RSUs, ∼90% pyramidal).