Figure 1 of the article by Birjandian et al. (2013) contained a minor error in the legend, which we hereby rectify. In the original figure legend panel (C) is described as two examples of whole cell patch recordings made from layer II and layer III neurons. But in fact panel (C) correspond to normalized change in ΔF/F vs. stimulus intensity from 8 recordings in the layer II. Panels (D,E) correspond to normalized change in ΔF/F vs. stimulus intensity from 8 recordings in DEn and Layer III but not layer II. We therefore would like to make a correction to Figures 1C–E.
Conflict of interest statement
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
References
- Birjandian Z., Narla C., Poulter M. O. (2013). Gain control of gamma frequency activation by a novel feed forward disinhibitory loop: implications for normal and epileptic neural activity. Front. Neural Circuits 7:183 10.3389/fncir.2013.00183 [DOI] [PMC free article] [PubMed] [Google Scholar]