Small Networks Encode Decision-Making in Primary Auditory Cortex

Abstract

Sensory detection tasks enhance representations of behaviorally meaningful stimuli in primary auditory cortex (A1). However, it remains unclear how A1 encodes decision-making. Neurons in A1 layer 2/3 (L2/3) show heterogeneous stimulus selectivity and complex anatomical connectivity, and receive input from prefrontal cortex. Thus, task-related modulation of activity in A1 L2/3 might differ across subpopulations. To study the neural coding of decision-making, we used two-photon imaging in A1 L2/3 of mice performing a tone-detection task. Neural responses to targets showed attentional gain and encoded behavioral choice. To characterize network representation of behavioral choice, we analyzed functional connectivity using Granger causality, pairwise noise correlations, and neural decoding. During task performance, small groups of four to five neurons became sparsely linked, locally clustered, and rostro-caudally oriented, while noise correlations both increased and decreased. Our results suggest that sensory-based decision-making involves small neural networks driven by the sum of sensory input, attentional gain, and behavioral choice.

Sensory and Behavioral Components of Neocortical Signal Flow in Discrimination Tasks with Short-term Memory

“…Compared to passive listening, sound-evoked activity in A1 may be enhanced or reduced in different auditory tasks (Francis et al, 2018;Jaramillo and Zador, 2011;Kato et al, 2015;Kuchibhotla et al, 2017;Otazu et al, 2009;Xin et al, 2019). Moreover, the causal involvement of A1 in auditory tasks remains controversial in view of conflicting results depending on task and inactivation method (Ohl et al, 1999;Jaramillo and Zador, 2011;Gimenez, Lorenc and Jaramillo, 2015;Kato, Gillet and Isaacson, 2015;Kuchibhotla et al, 2017;Talwar, Musial and Gerstein, 2017;Xin et al, 2019).…”

Functional Organization of Mouse Primary Auditory Cortex in adult C57BL/6 and F1 (CBAxC57) mice

“…We performed in vivo 2-photon imaging experiments in awake mice of C57 strain as well as CBA background (CBAxC57 F1) and quantified the functional responses across L2/3 and L4 of A1. We focused on young adult animals (~2-3 months old) which is the age frequently used in functional studies, especially those involving mouse behavior (Bandyopadhyay et al, 2010;Rothschild et al, 2010;Carcea et al, 2017;Guo et al, 2017;Kuchibhotla et al, 2017;Li et al, 2017;Meng et al, 2017;Resnik and Polley, 2017;Francis et al, 2018;Liu et al, 2019;Tischbirek et al, 2019). To compare strains we imaged the mid-frequency region of A1 (4-16kHz) which is also where mice are most sensitive to sound (Zheng et al, 1999).We find that A1 of CBA mice contains more neurons tuned to high frequencies and a higher number of significant tone-responses overall, including a higher proportion of off-responses (Liu et al, 2019).…”

Dynamic perceptual feature selectivity in primary somatosensory cortex upon reversal learning

“…First-order sensory areas encode lower-order stimulus features, such as textures coarseness (Chen et al, 2013a;Garion et al, 2014;Safaai et al, 2013), object orientation and direction (Hubel and Wiesel, 1959), and sound frequency (Stiebler et al, 1997), whereas more complex features and contextual aspects of a stimulus are encoded by higher-order cortices (Akrami et al, 2018;Felleman and Van Essen, 1991;Hwang et al, 2017;Pho et al, 2018). Nonetheless, the coding in primary sensory cortices can exhibit higher levels of complexity, expressing non-sensory-related signals such as attention (Francis et al, 2018), anticipation (Poort et al, 2015) and behavioral choice (Chen et al, 2015;Francis et al, 2018;Pho et al, 2018;Poort et al, 2015;Yang et al, 2016). Perceptual learning initially shapes the stimulus selectivity and response properties of primary sensory neurons, which may contribute to a reliable detection of particular features, and thereby improve perception (Goltstein et al, 2013;Pecka et al, 2014;Poort et al, 2015;Schoups et al, 2001).…”