Similarities in Response Non-Linearities in Macaque Lateral Prefrontal Cortex Visual Neurons During in Vivo and in Vitro Experiments. Implications for Normalization Models.

Visual neurons in many brain areas show non-linear response profiles as a function of the stimulus shown inside their receptive fields. These can be fit with different non-linear functions to obtain the tuning curve of the neuron for a particular feature. One example is the contrast response function, e.g., increases in the contrast of a stimulus inside a neuron’s receptive field produce changes in its response profile that can be fitted by a sigmoid function. Such properties have been attributed to lateral inhibition and normalization within a network of interconnected neurons. Here we test the hypothesis that non-linearities in response functions of single neurons during in vivorecordings can be at least in part attributed to their intrinsic (not network dependent) response properties. To address this issue, we first obtained response functions from single neuron recordings in the lateral prefrontal cortex (LPFC areas 8A/9/46) of two macaques to gratings of varying contrast inside their receptive fields. We then conducted patch clamp in vitrorecordings in slices extracted from the same LPFC area of 4 macaques using square current pulses of varying intensities that attempted to simulate increases in input strength when increasing contrast. In both datasets we convert spikes trains to firing rates over 250ms of stimulus presentation (in vivo) or pulse duration (in vitro) and fit the data with a sigmoid and a linear function. From 27 in vivo neurons, 52% were best fitted by the sigmoid and 48% were best fitted by a line. From 31 in vitro neurons 45% were best fitted by the sigmoid and 55% by a line. The proportions of neurons fitted with either function was not significantly different between areas (p>0.1, Chi-Square test) suggesting that non-linearities in the responses of visual neurons can be explained to a large degree by intrinsic cell properties.